You are on page 1of 66

# Fibonacci

PDF generated using the open source mwlib toolkit. See http://code.pediapress.com/ for more information. PDF generated at: Wed, 26 Jan 2011 19:05:46 UTC

Contents
Articles
Fibonacci BrahmaguptaFibonacci identity Carmichael's theorem Fibonacci coding Fibonacci cube Fibonacci heap Fibonacci polynomials Fibonacci prime Fibonacci pseudoprime Fibonacci retracement Fibonacci search technique Fibonacci triangle FibonacciSylvester expansion Lagged Fibonacci generator Negafibonacci NegaFibonacci coding Pisano period Reciprocal Fibonacci constant YoungFibonacci lattice The Fibonacci Association Fibonacci Quarterly Fibonacci numbers 1 4 6 7 10 13 17 18 19 22 24 25 26 30 32 34 36 39 40 42 43 44

References
Article Sources and Contributors Image Sources, Licenses and Contributors 61 63

Fibonacci

Fibonacci
Fibonacci
Nationality Italian Fields Mathematician

Knownfor Fibonacci number Fibonacci prime BrahmaguptaFibonacci identity Fibonacci polynomials Fibonacci pseudoprime Fibonacci word Reciprocal Fibonacci constant Introduction of digital notation to Europe Pisano period Practical number

Leonardo Pisano Bigollo (c. 1170 c. 1250)[1] also known as Leonardo of Pisa, Leonardo Pisano, Leonardo Bonacci, Leonardo Fibonacci, or, most commonly, simply Fibonacci, was an Italian mathematician, considered by some "the most talented western mathematician of the Middle Ages."[2] Fibonacci is best known to the modern world for [3] the spreading of the Hindu-Arabic numeral system in Europe, primarily through the publication in the early 13th century of his Book of Calculation, the Liber Abaci; and for a number sequence named after him known as the Fibonacci numbers, which he did not discover but used as an example in the Liber Abaci.[4]

Life
Leonardo Fibonacci was born around 1170 to Guglielmo Fibonacci, a wealthy Portrait by unknown artist Italian merchant. Guglielmo directed a trading post (by some accounts he was the consultant for Pisa) in Bugia, a port east of Algiers in the Almohad dynasty's sultanate in North Africa (now Bejaia, Algeria). As a young boy, Leonardo traveled with him to help; it was there he learned about the Hindu-Arabic numeral system.[5] Recognizing that arithmetic with Hindu-Arabic numerals is simpler and more efficient than with Roman numerals, Fibonacci traveled throughout the Mediterranean world to study under the leading Arab mathematicians of the time. Leonardo returned from his travels around 1200. In 1202, at age 32, he published what he had learned in Liber Abaci (Book of Abacus or Book of Calculation), and thereby popularized Hindu-Arabic numerals in Europe. Leonardo became an amicable guest of the Emperor Frederick II, who enjoyed mathematics and science. In 1240 the Republic of Pisa honored Leonardo, referred to as Leonardo Bigollo,[6] by granting him a salary. In the 19th century, a statue of Fibonacci was constructed and erected in Pisa. Today it is located in the western gallery of the Camposanto, historical cemetery on the Piazza dei Miracoli.[7]

Fibonacci

Liber Abaci
In the Liber Abaci (1202), Fibonacci introduces the so-called modus Indorum (method of the Indians), today known as Arabic numerals (Sigler 2003; Grimm 1973). The book advocated numeration with the digits 09 and place value. The book showed the practical importance of the new numeral system, using lattice multiplication and Egyptian fractions, by applying it to commercial bookkeeping, conversion of weights and measures, the calculation of interest, money-changing, and other applications. The book was well received throughout educated Europe and had a profound impact on European thought.

Fibonacci sequence
Liber Abaci also posed, and solved, a problem involving the growth of a population of rabbits based on idealized assumptions. The solution, generation by generation, was a sequence of numbers later known as Fibonacci numbers. The number sequence was known to Indian mathematicians as early as the 6th century,[8] [9] [10] but it was Fibonacci's Liber Abaci that introduced it to the West. In the Fibonacci sequence of numbers, each number is the sum of the previous two numbers, starting with 0 and 1. This sequence begins 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987 ... The higher up in the sequence, the closer two consecutive "Fibonacci numbers" of the sequence divided by each other will approach the golden ratio (approximately 1:1.618 or 0.618:1).

In popular culture
Fibonacci's name was adopted by a Los Angeles-based art rock group, The Fibonaccis, that recorded from 1981 to 1987. Stock traders frequently look to the "Fibonacci retracement" when predicting future share prices. A youthful Fibonacci is one of the main characters in the novel Crusade in Jeans (1973). He was left out of the 2006 movie version, however.

## Books written by Fibonacci

Liber Abaci (1202), a book on calculations (English translation by Laurence Sigler, Springer, 2002) Practica Geometriae (1220), a compendium on geometry and trigonometry. Flos (1225), solutions to problems posed by Johannes of Palermo Liber quadratorum, ("The Book of Squares") on Diophantine equations, dedicated to Emperor Frederick II. See in particular the BrahmaguptaFibonacci identity. Di minor guisa (on commercial arithmetic; lost) Commentary on Book X of Euclid's Elements (lost)

Notes
[1] "The Fibonacci Series - Biographies - Leonardo Fibonacci (ca.1175 - ca.1240)" (http:/ / library. thinkquest. org/ 27890/ biographies1. html). Library.thinkquest.org. . Retrieved 2010-08-02. [2] Howard Eves. An Introduction to the History of Mathematics. Brooks Cole, 1990: ISBN 0-03-029558-0 (6th ed.), p 261.

## 19th century statue of Fibonacci in Camposanto, Pisa.

[3] Leonardo Pisano - page 3: "Contributions to number theory" (http:/ / www. britannica. com/ eb/ article-4153/ Leonardo-Pisano). Encyclopdia Britannica Online, 2006. Retrieved 18 September 2006.

Fibonacci
[4] Parmanand Singh. "Acharya Hemachandra and the (so called) Fibonacci Numbers". Math. Ed. Siwan , 20(1):28-30, 1986. ISSN 0047-6269] [5] Dr R Knott: fibandphi (AT) ronknott DOT com. "Who was Fibonacci?" (http:/ / www. maths. surrey. ac. uk/ hosted-sites/ R. Knott/ Fibonacci/ fibBio. html). Maths.surrey.ac.uk. . Retrieved 2010-08-02. [6] See the incipit of Flos: "Incipit flos Leonardi bigolli pisani..." (quoted in the MS Word document Sources in Recreational Mathematics: An Annotated Bibliography (http:/ / www. g4g4. com/ MyCD5/ SOURCES/ SOURCE1. DOC) by David Singmaster, 18 March 2004 - emphasis added), in English: "Here starts 'the flower' by Leonardo the wanderer of Pisa..." The basic meanings of "bigollo" appear to be "good-for-nothing" and "traveler" (so it could be translated by "vagrant", "vagabond" or "tramp"). A. F. Horadam contends a connotation of "bigollo" is "absent-minded" (see first footnote of "Eight hundred years young" (http:/ / faculty. evansville. edu/ ck6/ bstud/ fibo. html)), which is also one of the connotations of the English word "wandering". The translation "the wanderer" in the quote above tries to combine the various connotations of the word "bigollo" in a single English word. [7] "Fibonacci's Statue in Pisa" (http:/ / www. epsilones. com/ documentos/ d-fibonacci. html#fibonacci-ingles). Epsilones.com. . Retrieved 2010-08-02. [8] Susantha Goonatilake (1998). Toward a Global Science (http:/ / books. google. com/ ?id=SI5ip95BbgEC& pg=PA126& dq=Virahanka+ Fibonacci). Indiana University Press. p.126. ISBN9780253333889. . [9] Donald Knuth (2006). The Art of Computer Programming: Generating All TreesHistory of Combinatorial Generation; Volume 4 (http:/ / books. google. com/ ?id=56LNfE2QGtYC& pg=PA50& dq=rhythms). Addison-Wesley. p.50. ISBN9780321335708. . [10] Rachel W. Hall. Math for poets and drummers (http:/ / www. sju. edu/ ~rhall/ mathforpoets. pdf). Math Horizons 15 (2008) 10-11.

References
Goetzmann, William N. and Rouwenhorst, K.Geert, The Origins of Value: The Financial Innovations That Created Modern Capital Markets (2005, Oxford University Press Inc, USA), ISBN 0195175719. Grimm, R. E., "The Autobiography of Leonardo Pisano", Fibonacci Quarterly, Vol. 11, No. 1, February 1973, pp. 99-104. A. F. Horadam, "Eight hundred years young," The Australian Mathematics Teacher 31 (1975) 123-134.

Fibonacci Biography (http://library.thinkquest.org/27890/biographies1.html) Who was Fibonacci? (http://www.mcs.surrey.ac.uk/Personal/R.Knott/Fibonacci/fibBio.html) by Ron Knott. Goetzmann, William N., Fibonacci and the Financial Revolution (http://papers.ssrn.com/sol3/papers. cfm?abstract_id=461740) (October 23, 2003), Yale School of Management International Center for Finance Working Paper No. 03-28 Fibonacci (http://mathdl.maa.org/convergence/1/?pa=content&sa=viewDocument&nodeId=630& bodyId=1002) at Convergence (http://mathdl.maa.org/convergence/1/) O'Connor, John J.; Robertson, Edmund F., "Leonardo Pisano Fibonacci" (http://www-history.mcs.st-andrews. ac.uk/Biographies/Fibonacci.html), MacTutor History of Mathematics archive, University of St Andrews.

BrahmaguptaFibonacci identity

BrahmaguptaFibonacci identity
In algebra, Brahmagupta's identity, also called Fibonacci's identity, implies that the product of two sums of two squares is itself a sum of two squares. In other words, the set of all sums of two squares is closed under multiplication. Specifically:

For example,

The identity is a special case (n=2) of Lagrange's identity, and is first found in Diophantus. Brahmagupta proved and used a more general identity, equivalent to

## is closed under multiplication.

Both (1) and (2) can be verified by expanding each side of the equation. Also, (2) can be obtained from (1), or (1) from (2), by changing b tob. This identity holds in both the ring of integers and the ring of rational numbers, and more generally in any commutative ring. In the integer case this identity finds applications in number theory for example when used in conjunction with one of Fermat's theorems it proves that the product of a square and any number of primes of the form 4n+1 is also a sum of two squares.

History
The identity is first found in Diophantus's Arithmetica (III, 19). The identity was rediscovered by Brahmagupta (598668), an Indian mathematician and astronomer, who generalized it. His Brahmasphutasiddhanta was translated from Sanskrit into Arabic by Mohammad al-Fazari, and was subsequently translated into Latin in 1126.[1] The identity later appeared in Fibonacci's Book of Squares in 1225.

Related identities
Euler's four-square identity is an analogous identity involving four squares instead of two that is related to quaternions. There is a similar eight-square identity derived from the Cayley numbers which has connections to Bott periodicity.

BrahmaguptaFibonacci identity

## Relation to complex numbers

If a, b, c, and d are real numbers, this identity is equivalent to the multiplication property for absolute values of complex numbers namely that:

since

## Interpretation via norms

In the case that the variables a, b, c, and d are rational numbers, the identity may be interpreted as the statement that the norm in the field Q(i) is multiplicative. That is, we have

and also

## Application to Pell's equation

It its original context, Brahmagupta applied his discovery to the solution of Pell's equation, namely x2Ny2=1. Using the identity in the more general form he was able to "compose" triples (x1,y1,k1) and (x2,y2,k2) that were solutions of x2Ny2=k, to generate the new triple Not only did this give a way to generate infinitely many solutions to x2Ny2=1 starting with one solution, but also, by dividing such a composition by k1k2, integer or "nearly integer" solutions could often be obtained. The general method for solving the Pell equation given by Bhaskara II in 1150, namely the chakravala (cyclic) method, was also based on this identity.[2]

BrahmaguptaFibonacci identity

References
[1] George G. Joseph (2000). The Crest of the Peacock, p. 306. Princeton University Press. ISBN 0691006598. [2] John Stillwell (2002), Mathematics and its history (http:/ / books. google. com/ books?id=WNjRrqTm62QC& pg=PA72) (2 ed.), Springer, pp.7276, ISBN9780387953366,

Brahmagupta's identity at [[PlanetMath (http://planetmath.org/encyclopedia/BrahmaguptasIdentity.html)]] Brahmagupta Identity (http://mathworld.wolfram.com/BrahmaguptaIdentity.html) on MathWorld Brahmagupta-Fibonacci identity (http://www.pballew.net/fiboiden.html) A Collection of Algebraic Identities (http://sites.google.com/site/tpiezas/005b/)

Carmichael's theorem
This article refers to Carmichael's theorem about Fibonacci numbers. Carmichael's theorem may also refer to the recursive definition of the Carmichael function. Carmichael's theorem, named after the American mathematician R.D. Carmichael, states that for n greater than 12, the nth Fibonacci number F(n) has at least one prime factor that is not a factor of any earlier Fibonacci number. The only exceptions for n up to 12 are: F(1)=1 and F(2)=1, which have no prime factors F(6)=8 whose only prime factor is 2 (which is F(3)) F(12)=144 whose only prime factors are 2 (which is F(3)) and 3 (which is F(4)) If a prime p is a factor of F(n) and not a factor of any F(m) with m < n then p is called a characteristic factor or a primitive divisor of F(n). Carmichael's theorem says that every Fibonacci number, apart from the exceptions listed above, has at least one characteristic factor.

References
Carmichael, R. D. (1913), "On the numerical factors of the arithmetic forms n+n" [1], Annals of Mathematics (Annals of Mathematics) 15 (1/4): 3070, doi:10.2307/1967797. Knott, R., Fibonacci numbers and special prime factors [2], Fibonacci Numbers and the Golden Section [3]. Yabuta, M. (2001), "A simple proof of Carmichael's theorem on primitive divisors", Fibonacci Quarterly 39: 439443.

References
[1] http:/ / jstor. org/ stable/ 1967797 [2] http:/ / www. mcs. surrey. ac. uk/ Personal/ R. Knott/ Fibonacci/ fibmaths. html#primefactor [3] http:/ / www. mcs. surrey. ac. uk/ Personal/ R. Knott/ Fibonacci/ fib. html

Fibonacci coding

Fibonacci coding
Numeral systems by culture Hindu-Arabic numerals Western Arabic Eastern Arabic Indian family Burmese Khmer Mongolian Thai

East Asian numerals Chinese Japanese Suzhou Korean Vietnamese Counting rods Alphabetic numerals Abjad Armenian ryabhaa Cyrillic Ge'ez Greek (Ionian) Hebrew

Other systems Aegean Attic Babylonian Brahmi Egyptian Etruscan Inuit Mayan Quipu Roman Sumerian Urnfield

List of numeral system topics Positional systems by base Decimal (10) 1, 2, 3, 4, 5, 6, 8, 12, 16, 20, 30, 36, 60 more

In mathematics, Fibonacci coding is a universal code which encodes positive integers into binary code words. All tokens end with "11" and have no "11" before the end. For a number , if , and represent the digits of the coded form of . then we have:

where F(i) is the ith Fibonacci number. No two adjacent coefficients d(i) can be 1. It can be shown that such a coding is unique, and in the code "11" never appears anywhere but the end. The code begins as follows:

Fibonacci coding

Symbol 1 2 3 4 5 6 7 8

Fibonacci representation 1 2 4 5 8 9 10 16

## Fibonacci code 11 011 0011 1011 00011 10011 01011 000011

The Fibonacci code is closely related to the Zeckendorf representation, a positional numeral system that uses Zeckendorf's theorem and has the property that no number has a representation with consecutive 1's. The Zeckendorf representation for a particular integer is exactly that of the integer's Fibonacci representation, except with the order of its digits reversed and an additional "1" appended to the end. To encode an integer X: 1. Find the largest Fibonacci number equal to or less than X; subtract this number from X, keeping track of the remainder. 2. If the number we subtracted was the Nth unique Fibonacci number, put a one in the Nth digit of our output. 3. Repeat the previous steps, substituting our remainder for X, until we reach a remainder of 0. 4. Place a one after the last naturally-occurring one in our output. To decode a token in the code, remove the last "1", assign the remaining bits the values 1,2,3,5,8,13... (the Fibonacci numbers), and add the "1" bits.

## Comparison with other universal codes

Fibonacci coding has a useful property that sometimes makes it attractive in comparison to other universal codes: it is an example of a self-synchronizing code, making it easier to recover data from a damaged stream. With most other universal codes, if a single bit is altered, none of the data that comes after it will be correctly read. With Fibonacci coding, on the other hand, a changed bit may cause one token to be read as two, or cause two tokens to be read incorrectly as one, but reading a "0" from the stream will stop the errors from propagating further. Since the only stream that has no "0" in it is a stream of "11" tokens, the total edit distance between a stream damaged by a single bit error and the original stream is at most three. This approach - encoding using sequence of symbols, in which some patterns (like "11") are forbidden, can be freely generalized[1].

Example
The following table shows that the number 65 is represented in Fibonacci coding as 0100100011, since 65 = 2 + 8 + 55. The first two Fibonacci numbers (0 and 1) are not used, and an additional 1 is always appended.

Fibonacci coding

13

21

34

55

A method to encode any integer is shown in the following Python program. def encode_fib(n): # Return string with Fibonacci encoding for n (n >= 1). result = "" if n >= 1: a = 1 b = 1 c = a + b # next Fibonacci number fibs = [b] # list of Fibonacci numbers, starting with F(2), each <= n while n >= c: fibs.append(c) # add next Fibonacci number to end of list a = b b = c c = a + b result = "1" # extra "1" at end for fibnum in reversed(fibs): if n >= fibnum: n = n - fibnum result = "1" + result else: result = "0" + result return result print encode_fib(65) # displays "0100100011"

References
Fraenkel A, Klein S. (1996). "Robust universal complete codes for transmission and compression" [2]. Discrete Applied Mathematics 64: 3155.

Stakhov, A. P. (2009). The Mathematics of Harmony: From Euclid to Contemporary Mathematics and Computer Science. Singapore: World Scientific Publishing.

References
[1] http:/ / aps. arxiv. org/ pdf/ 0710. 3861 [2] http:/ / citeseerx. ist. psu. edu/ viewdoc/ summary?doi=10. 1. 1. 37. 3064

Fibonacci cube

10

Fibonacci cube
The Fibonacci cubes or Fibonacci networks are a family of undirected graphs with properties similar to those of hypercube graphs, but with a Fibonacci number of vertices, studied in graph-theoretic mathematics. The Fibonacci cube may be defined in terms of Fibonacci codes and Hamming distance, independent sets of vertices in path graphs, or via distributive lattices. Although the lattice definition is older, Fibonacci cubes were first explicitly defined as graphs by Hsu (1993) in the context of interconnection topologies for connecting parallel or distributed systems. They have also been applied in chemical graph theory.

## The Fibonacci cube of order 6

Definition
Like the hypercube graph, the vertices of the Fibonacci cube of order n may be labeled with bitstrings of length n, in such a way that two vertices are adjacent whenever their labels differ in a single bit. However, in a Fibonacci cube, the only labels that are allowed Fibonacci cubes as subgraphs of hypercubes are bitstrings with no two consecutive 1 bits. There are Fn+2 labels labels possible, where Fn denotes the nth Fibonacci number, and therefore there are Fn+2 vertices in the Fibonacci cube of order n. The nodes of such a network may be assigned consecutive integers from 0 to Fn+2; the bitstrings corresponding to these numbers are given by their Zeckendorf representations.

Algebraic structure
The Fibonacci cube of order n is the simplex graph of the complement graph of an n-vertex path graph. That is, each vertex in the Fibonacci cube represents a clique in the path complement graph, or equivalently an independent set in the path itself; two Fibonacci cube vertices are adjacent if the cliques or independent sets that they represent differ by the addition or removal of a single element. Therefore, like other simplex graphs, Fibonacci cubes are median graphs.[1] The median of any three vertices in a Fibonacci cube may be found by computing the bitwise majority function of the three labels; if each of the three labels has no two consecutive 1 bits, the same is true of their majority.

Fibonacci cube The Fibonacci cube is also the graph of a distributive lattice that may be obtained via Birkhoff's representation theorem from a zigzag poset, a partially ordered set defined by an alternating sequence of order relations a < b > c < d > e < f > ...[2] There is also an alternative graph-theoretic description of the same lattice: the independent sets of any bipartite graph may be given a partial order in which one independent set is less than another if they differ by removing elements from one side of the bipartition and adding elements to the other side of the bipartition; with this order, the independent sets form a distributive lattice,[3] and applying this construction to a path graph results in the lattice associated with the Fibonacci cube.

11

## Properties and algorithms

The Fibonacci cube of order n may be partitioned into a Fibonacci cube of order n1 (the nodes with labels beginning with a 0 bit) and a Fibonacci cube of order n2 (the nodes with labels beginning with a 1 bit). Every Fibonacci cube has a Hamiltonian path. More specifically, there exists a path that obeys the partition described above: it visits the nodes with first bit 0 and the nodes with first bit 1 in two contiguous subsequences. Within these two subsequences, the path can be constructed recursively by the same rule, linking the two subsequences at the ends of the subsequences at which the second bit is 0. Thus, e.g., in the Fibonacci cube of order 4, the sequence constructed in this way is (0100-0101-0001-0000-0010)-(1010-1000-1001), where the parentheses demark the subsequences within the two subgraphs of the partition. Fibonacci cubes with an even number of nodes greater than two have a Hamiltonian cycle.[4] Munarini & Salvi (2002) investigate the radius and independence number of Fibonacci cubes. Taranenko & Vesel (2007) showed that it is possible to test whether a graph is a Fibonacci cube in time near-linear in its size.

Applications
Hsu (1993) suggested using Fibonacci cubes as a network topology in parallel computing. As a communications network, the Fibonacci cube has beneficial properties similar to those of the hypercube: the number of incident edges per vertex is at most n/2 and the diameter of the network is at most n, both proportional to the logarithm of the number of vertices, and the ability of the network to be partitioned into smaller networks of the same type allows it to be split among multiple parallel computation tasks.[4] Fibonacci cubes also support efficient protocols for routing and broadcasting in distributed computations.[5] Klavar & igert (2005) apply Fibonacci cubes in chemical graph theory as a description of the family of perfect matchings of certain molecular graphs; this description allows them to characterize the resonance structure of the molecules.

Related graphs
Generalized Fibonacci cubes were presented by Hsu & Chung (1993) based on the k-th order Fibonacci numbers, which were later further extended to become the Linear Recursive Networks by Hsu, Chung & Das (1997) based on more general forms of linear recursions. Wu (1997) modified the second order Fibonacci cubes based on different initial conditions. Another related graph is the Lucas cube, a graph with a Lucas number of vertices defined from the Fibonacci cube by forbidding a 1 bit in both the first and last positions of each bitstring; Ded, Torri & Salvi (2002) investigated the coloring properties of both Fibonacci cubes and Lucas cubes.

Fibonacci cube

12

Notes
[1] Klavar (2005). [2] Gansner (1982) calls the fact that this lattice has a Fibonacci number of elements a well known fact, while Stanley (1986) asks for a description of it in an exercise. See also Hft & Hft (1985), Beck (1990), and Salvi & Salvi (2008). [3] Propp (1997). [4] Cong, Zheng & Sharma (1993). [5] Hsu (1993); Stojmenovic 1998.

References
Beck, Istvn (1990), "Partial orders and the Fibonacci numbers", Fibonacci Quarterly 28 (2): 172174, MR1051291. Cong, B.; Zheng, S. Q.; Sharma, S. (1993), "On simulations of linear arrays, rings and 2D meshes on Fibonacci cube networks", Proc. 7th Int. Parallel Processing Symposium, pp.748751, doi:10.1109/IPPS.1993.262788. Ded, Ernesto; Torri, Damiano; Salvi, Norma Zagaglia (2002), "The observability of the Fibonacci and the Lucas cubes", Discrete Mathematics 255 (13): 5563, doi:10.1016/S0012-365X(01)00387-9. Gansner, Emden R. (1982), "On the lattice of order ideals of an up-down poset", Discrete Mathematics 39 (2): 113122, doi:10.1016/0012-365X(82)90134-0, MR675856. Hft, Hartmut; Hft, Margret (1985), "A Fibonacci sequence of distributive lattices", Fibonacci Quarterly 23 (3): 232237, MR806293. Hsu, W.-J. (1993), "Fibonacci cubes: a new interconnection topology", IEEE Transactions on Parallel and Distributed Systems 4 (1): 312, doi:10.1109/71.205649. Hsu, W.-J.; Chung, M. J. (1993), "Generalized Fibonacci cubes", 1993 International Conference on Parallel Processing - ICPP'93, 1, pp.299302, doi:10.1109/ICPP.1993.95. Hsu, W.-J.; Page, C. V.; Liu, J.-S. (1993), "Fibonacci cubes: a class of self-similar graphs", Fibonacci Quarterly 31 (1): 6572. Hsu, W.-J.; Chung, M. J.; Das, A. (1997), "Linear recursive networks and their applications in distributed systems", IEEE Transactions on Parallel and Distributed Systems 8 (7): 673680, doi:10.1109/71.598343. Klavar, Sandi (2005), "On median nature and enumerative properties of Fibonacci-like cubes", Discrete Mathematics 299 (13): 145153, doi:10.1016/j.disc.2004.02.023. Klavar, Sandi; igert, Petra (2005), "Fibonacci cubes are the resonance graphs of Fibonaccenes" (http:// www-mat.pfmb.uni-mb.si/personal/klavzar/preprints/Fibonacci6.ps), Fibonacci Quarterly 43 (3): 269276. Munarini, Emanuele; Salvi, Norma Zagaglia (2002), "Structural and enumerative properties of the Fibonacci cubes", Discrete Mathematics 255 (13): 317324, doi:10.1016/S0012-365X(01)00407-1. Propp, James (1997), "Generating random elements of finite distributive lattices", Electronic Journal of Combinatorics 4 (2): R15, arXiv:math.CO/9801066. Salvi, Rodolfo; Salvi, Norma Zagaglia (2008), "Alternating unimodal sequences of Whitney numbers", Ars Combinatoria 87: 105117, MR2414008. Stanley, Richard P. (1986), Enumerative Combinatorics, Wadsworth, Inc. Exercise 3.23a, page 157. Stojmenovic, Ivan (1998), "Optimal deadlock-free routing and broadcasting on Fibonacci cube networks" (http:// www.site.uottawa.ca/~ivan/F51-fibonacciCube.PDF), Utilitas Mathematica 53: 159166. Taranenko, A.; Vesel, A. (2007), "Fast recognition of Fibonacci cubes", Algorithmica 49 (2): 8193, doi:10.1007/s00453-007-9026-5. Wu, Jie (1997), "Extended Fibonacci cubes", IEEE Transactions on Parallel and Distributed Systems 8 (12): 12031210, doi:10.1109/71.640012.

Fibonacci heap

13

Fibonacci heap
In computer science, a Fibonacci heap is a heap data structure consisting of a collection of trees. It has a better amortized running time than a binomial heap. Fibonacci heaps were developed by Michael L. Fredman and Robert E. Tarjan in 1984 and first published in a scientific journal in 1987. The name of Fibonacci heap comes from Fibonacci numbers which are used in the running time analysis. Operations insert, find minimum, decrease key, and merge (union) work in constant amortized time. Operations delete and delete minimum work in O(log n) amortized time. This means that starting from an empty data structure, any sequence of a operations from the first group and b operations from the second group would take O(a+blogn) time. In a binomial heap such a sequence of operations would take O((a+b)log (n)) time. A Fibonacci heap is thus better than a binomial heap when b is asymptotically smaller than a. Using Fibonacci heaps for priority queues improves the asymptotic running time of important algorithms, such as Dijkstra's algorithm for computing Shortest paths.

Structure
A Fibonacci heap is a collection of trees satisfying the minimum-heap property, that is, the key of a child is always greater than or equal to the key of the parent. This implies that the minimum key is always at the root of one of the trees. Compared with binomial heaps, the structure of a Fibonacci heap is more flexible. The trees do not have a prescribed shape and in the extreme case the heap can have every element in a separate tree or a single tree of depth n. This flexibility allows some operations to be executed in a "lazy" manner, postponing the work for later operations. For example merging heaps is done simply by concatenating the two lists of trees, and operation decrease key sometimes cuts a node from its parent and forms a new tree.

Figure 1. Example of a Fibonacci heap. It has three trees of degrees 0, 1 and 3. Three vertices are marked (shown in blue). Therefore the potential of the heap is 9.

However at some point some order needs to be introduced to the heap to achieve the desired running time. In particular, degrees of nodes (here degree means the number of children) are kept quite low: every node has degree at most O(log n) and the size of a subtree rooted in a node of degree k is at least Fk+2, where Fk is the kth Fibonacci number. This is achieved by the rule that we can cut at most one child of each non-root node. When a second child is cut, the node itself needs to be cut from its parent and becomes the root of a new tree (see Proof of degree bounds, below). The number of trees is decreased in the operation delete minimum, where trees are linked together. As a result of a relaxed structure, some operations can take a long time while others are done very quickly. In the amortized running time analysis we pretend that very fast operations take a little bit longer than they actually do. This additional time is then later subtracted from the actual running time of slow operations. The amount of time saved for later use is measured at any given moment by a potential function. The potential of a Fibonacci heap is given by Potential = t + 2m where t is the number of trees in the Fibonacci heap, and m is the number of marked nodes. A node is marked if at least one of its children was cut since this node was made a child of another node (all roots are unmarked).

Fibonacci heap Thus, the root of each tree in a heap has one unit of time stored. This unit of time can be used later to link this tree with another tree at amortized time 0. Also, each marked node has two units of time stored. One can be used to cut the node from its parent. If this happens, the node becomes a root and the second unit of time will remain stored in it as in any other root.

14

Implementation of operations
To allow fast deletion and concatenation, the roots of all trees are linked using a circular, doubly linked list. The children of each node are also linked using such a list. For each node, we maintain its number of children and whether the node is marked. Moreover we maintain a pointer to the root containing the minimum key. Operation find minimum is now trivial because we keep the pointer to the node containing it. It does not change the potential of the heap, therefore both actual and amortized cost is constant. As mentioned above, merge is implemented simply by concatenating the lists of tree roots of the two heaps. This can be done in constant time and the potential does not change, leading again to constant amortized time. Operation insert works by creating a new heap with one element and doing merge. This takes constant time, and the potential increases by one, because the number of trees increases. The amortized cost is thus still constant. Operation extract minimum (same as delete minimum) operates in three phases. First we take the root containing the minimum element and remove it. Its children will become roots of new trees. If the number of children was d, it takes time O(d) to process all new roots and the potential increases by d-1. Therefore the amortized running time of this phase is O(d) = O(log n).

Fibonacci heap from Figure 1 after first phase of extract minimum. Node with key 1 (the minimum) was deleted and its children were added as separate trees.

In the third phase we check each of the remaining roots and find the minimum. This takes O(log n) time and the potential does not change. The overall amortized running time of extract minimum is therefore O(log n).

Fibonacci heap from Figure 1 after extract minimum is completed. First, nodes 3 and 6 are linked together. Then the result is linked with tree rooted at node 2. Finally, the new minimum is found.

However to complete the extract minimum operation, we need to update the pointer to the root with minimum key. Unfortunately there may be up to n roots we need to check. In the second phase we therefore decrease the number of roots by successively linking together roots of the same degree. When two roots u and v have the same degree, we make one of them a child of the other so that the one with smaller key remains the root. Its degree will increase by one. This is repeated until every root has a different degree. To find trees of the same degree efficiently we use an array of length O(log n) in which we keep a pointer to one root of each degree. When a second root is found of the same degree, the two are linked and the array is updated. The actual running time is O(log n + m) where m is the number of roots at the beginning of the second phase. At the end we will have at most O(log n) roots (because each has a different degree). Therefore the potential decreases by at least m-O(log n) and the amortized running time is O(log n).

Fibonacci heap

15

Operation decrease key will take the node, decrease the key and if the heap property becomes violated (the new key is smaller than the key of the parent), the node is cut from its parent. If the parent is not a root, it is marked. If it has been marked already, it is cut as well and its parent is marked. We continue upwards until we reach either the root or an unmarked node. In the Fibonacci heap from Figure 1 after decreasing key of node 9 to 0. This node as well as its two marked ancestors are cut from the tree process we create some number, say k, of new trees. rooted at 1 and placed as new roots. Each of these new trees except possibly the first one was marked originally but as a root it will become unmarked. One node can become marked. Therefore the potential decreases by at least k2. The actual time to perform the cutting was O(k), therefore the amortized running time is constant. Finally, operation delete can be implemented simply by decreasing the key of the element to be deleted to minus infinity, thus turning it into the minimum of the whole heap. Then we call extract minimum to remove it. The amortized running time of this operation is O(log n).

## Proof of degree bounds

The amortized performance of a Fibonacci heap depends on the degree (number of children) of any tree root being O(log n), where n is the size of the heap. Here we show that the size of the (sub)tree rooted at any node x of degree d in the heap must have size at least Fd+2, where Fk is the kth Fibonacci number. The degree bound follows from this and the fact (easily proved by induction) that for all integers , where . (We then have , and taking the log to base of both sides gives

as required.) Consider any node x somewhere in the heap (x need not be the root of one of the main trees). Define size(x) to be the size of the tree rooted at x (the number of descendants of x, including x itself). We prove by induction on the height of x (the length of a longest simple path from x to a descendant leaf), that size(x)Fd+2, where d is the degree of x. Base case: If x has height 0, then d=0, and size(x)=1=F2. Inductive case: Suppose x has positive height and degree d>0. Let y1, y2, ..., yd be the children of x, indexed in order of the times they were most recently made children of x (y1 being the earliest and yd the latest), and let c1, c2, ..., cd be their respective degrees. We claim that cii-2 for each i with 2id: Just before yi was made a child x, y1,...,yi-1 were already children of x, and so x had degree at least i-1 at that time. Since trees are combined only when the degrees of their roots are equal, it must have been that yi also had degree at least i-1 at the time it became a child of x. From that time to the present, yi can only have lost at most one child (as guaranteed by the marking process), and so its current degree ci is at least i-2. This proves the claim. Since the heights of all the yi are strictly less than that of x, we can apply the inductive hypothesis to them to get size(yi)Fci+2F(i-2)+2=Fi. The nodes x and y1 each contribute at least 1 to size(x), and so we have

for any

## , which gives the desired lower bound on

Fibonacci heap

16

Worst case
Although the total running time of a sequence of operations starting with an empty structure is bounded by the bounds given above, some (very few) operations in the sequence can take very long to complete (in particular delete and delete minimum have linear running time in the worst case). For this reason Fibonacci heaps and other amortized data structures may not be appropriate for real-time systems.

## Summary of running times

Unsorted Linked List insert accessmin O(1) O(n) Balanced binary Tree O(log n) O(log n) or O(1) (**) O(log n) O(log n) O(log n) O(m log(n+m)) (Min-)Heap Binomial Heap O(log n) O(1) Fibonacci Heap O(1) O(1) Brodal [1] Queue O(1) O(1) Pseudo [2] Queue O(1) O(1) Pairing heap O(1) O(1)

O(log n) O(1)

deletemin

O(n)

## decreasekey O(1) delete merge O(n) O(1)

(*)Amortized time (**)With trivial modification to the conventional Binary tree (***)Where n is the size of the larger heap

References
[1] (http:/ / citeseerx. ist. psu. edu/ viewdoc/ summary?doi=10. 1. 1. 43. 8133) [2] (http:/ / www. springerlink. com/ content/ p5571w5jh1k7l2k0/ )

Fredman M. L. & Tarjan R. E. (1987). Fibonacci heaps and their uses in improved network optimization algorithms. (http://www.cl.cam.ac.uk/~sos22/supervise/dsaa/fib_heaps.pdf) Journal of the ACM 34(3), 596-615. Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest, and Clifford Stein. Introduction to Algorithms, Second Edition. MIT Press and McGraw-Hill, 2001. ISBN 0-262-03293-7. Chapter 20: Fibonacci Heaps, pp.476497. Brodal, G. S. 1996. Worst-case efficient priority queues. (http://portal.acm.org/citation.cfm?id=313883) In Proceedings of the Seventh Annual ACM-SIAM Symposium on Discrete Algorithms (Atlanta, Georgia, United States, January 28 - 30, 1996). Symposium on Discrete Algorithms. Society for Industrial and Applied Mathematics, Philadelphia, PA, 52-58.

Java applet simulation of a Fibonacci heap (http://www.cs.yorku.ca/~aaw/Jason/FibonacciHeapAnimation. html) C implementation of Fibonacci heap (http://resnet.uoregon.edu/~gurney_j/jmpc/fib.html) De-recursived and memory efficient C implementation of Fibonacci heap (http://www.labri.fr/perso/pelegrin/ code/#fibonacci) (free/libre software, CeCILL-B license (http://www.cecill.info/licences/ Licence_CeCILL-B_V1-en.html)) C++ template Fibonacci heap, with demonstration (http://ideone.com/9jYnv) Ruby implementation of the Fibonacci heap (with tests) (http://github.com/evansenter/f_heap)

## Fibonacci heap Pseudocode of the Fibonacci heap algorithm (http://www.cs.princeton.edu/~wayne/cs423/fibonacci/ FibonacciHeapAlgorithm.html)

17

Fibonacci polynomials
In mathematics, the Fibonacci polynomials are a polynomial sequence which can be considered as a generalisation of the Fibonacci numbers.

Definition
These polynomials are defined by a recurrence relation:

Properties
The first few Fibonacci polynomials are:

The Fibonacci numbers are recovered by evaluating the polynomials at x=1. The degree of Fn is n1. The ordinary generating function for the sequence is

References
Hoggatt, V.E., jun.; Bicknell,e (1973). "Roots of Fibonacci polynomials.". Fibonacci Quarterly 11: 271274. ISSN0015-0517. Ricci, Paolo Emilio (1995). "Generalized Lucas polynomials and Fibonacci polynomials". Riv. Mat. Univ. Parma, V. Ser. 4: 137146.

Weisstein, Eric W., "Fibonacci Polynomial [1]" from MathWorld.

Fibonacci polynomials

18

References
[1] http:/ / mathworld. wolfram. com/ FibonacciPolynomial. html

Fibonacci prime
A Fibonacci prime is a Fibonacci number that is prime, a type of integer sequence prime. The first Fibonacci primes are (sequence A005478 [1] in OEIS): 2, 3, 5, 13, 89, 233, 1597, 28657, 514229, 433494437, 2971215073, ....

## Known Fibonacci primes

It is not known if there are infinitely many Fibonacci primes. The first 33 are Fn for the n values (sequence A001605 [2] in OEIS): 3, 4, 5, 7, 11, 13, 17, 23, 29, 43, 47, 83, 131, 137, 359, 431, 433, 449, 509, 569, 571, 2971, 4723, 5387, 9311, 9677, 14431, 25561, 30757, 35999, 37511, 50833, 81839. In addition to these proven Fibonacci primes, there have been found probable primes for n = 104911, 130021, 148091, 201107, 397379, 433781, 590041, 593689, 604711, 931517, 1049897, 1285607, 1636007, 1803059, 1968721.[3] Except for the case n = 4, all Fibonacci primes have a prime index, but not all prime indexes are a Fibonacci prime. Fp is prime for 8 out of the first 10 primes p; the exceptions are F2 = 1 and F19 = 4181 = 37 113. However, Fibonacci primes become rarer as the index increases. Fp is prime for only 25 of the 1,229 primes p below 10,000.[4] As of November 2009, the largest known certain Fibonacci prime is F81839, with 17103 digits. It was proved prime by David Broadhurst and Bouk de Water in 2001.[5] [6] The largest known probable Fibonacci prime is F1968721. It has 411439 digits and was found by Henri Lifchitz in 2009.[3]

## Divisibility of Fibonacci numbers

Fibonacci numbers that have a prime index p do not share any common divisors greater than 1 with the preceding Fibonacci numbers, due to the identity GCD(Fn, Fm) = FGCD(n,m).[7] (This implies the infinitude of primes.) For n3, Fn divides Fm iff n divides m.[8] If we suppose that m, is a prime number p from the identity above, and n is less than p, then it is clear that Fp, cannot share any common divisors with the preceding Fibonacci numbers. GCD(Fp, Fn) = FGCD(p,n) = F1 = 1 Carmichael's theorem states that every Fibonacci number (except for 1, 8 and 144) has at least one unique prime factor that has not been a factor of the preceding Fibonacci numbers.

Fibonacci prime

19

References
[1] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa005478 [2] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa001605 [3] PRP Top Records, Search for : F(n) (http:/ / www. primenumbers. net/ prptop/ searchform. php?form=F(n)& action=Search). Retrieved 2009-11-21. [4] Sloane's A005478 (http:/ / en. wikipedia. org/ wiki/ Oeis:a005478), A001605 (http:/ / en. wikipedia. org/ wiki/ Oeis:a001605) [5] Number Theory Archives announcement by David Broadhurst and Bouk de Water (http:/ / listserv. nodak. edu/ cgi-bin/ wa. exe?A2=ind0104& L=nmbrthry& P=R1807& D=0) [6] Chris Caldwell, The Top Twenty: Fibonacci Number (http:/ / primes. utm. edu/ top20/ page. php?id=39) from the Prime Pages. Retrieved 2009-11-21. [7] Paulo Ribenboim, My Numbers, My Friends, Springer-Verlag 2000 [8] Wells 1986, p.65

Weisstein, Eric W., " Fibonacci Prime (http://mathworld.wolfram.com/FibonacciPrime.html)" from MathWorld. R. Knott Fibonacci primes (http://www.mcs.surrey.ac.uk/Personal/R.Knott/Fibonacci/fibmaths. html#fibprimes) Caldwell, Chris. Fibonacci number (http://primes.utm.edu/glossary/page.php/FibonacciNumber.html), Fibonacci prime (http://primes.utm.edu/glossary/page.php?sort=FibonacciPrime), and Record Fibonacci primes (http://primes.utm.edu/top20/page.php?id=39) at the Prime Pages Small parallel Haskell program to find probable Fibonacci primes at haskell.org (http://www.haskell.org/ haskellwiki/Fibonacci_primes_in_parallel)

Fibonacci pseudoprime
In number theory, the classes of Lucas pseudoprime and Fibonacci pseudoprime comprise sequences of composite integers that passes certain tests that all primes pass: in this case, criteria relative to some Lucas sequence.

Definition
Given two integer parameters P and Q which satisfy

the Lucas sequences of the first and second kind, Un(P,Q) and Vn(P,Q) respectively, are defined by the recurrence relations

and

We can write

Fibonacci pseudoprime where a and b are roots of the auxiliary polynomial X2 - P X + Q, of discriminant D. If p is an odd prime number then Vp is congruent to P modulo p. and if the Jacobi symbol , then p is a factor of Up-.

20

Lucas pseudoprimes
A Lucas pseudoprime is a composite number n for which n is a factor of Un-. (Riesel adds the condition that D should be 1.) In the specific case of the Fibonacci sequence, where P = 1, Q = -1 and D = 5, the first Lucas pseudoprimes are 323 and 377; multiple of 377. A strong Lucas pseudoprime is an odd composite number n with (n,D)=1, and n-=2rs with s odd, satisfying one of the conditions n divides Us n divides V2js for some j < r. A strong Lucas pseudoprime is also a Lucas pseudoprime. An extra strong Lucas pseudoprime is a strong Lucas pseudoprime for a set of parameters (P,Q) where Q = 1, satisfying one of slightly modified conditions n divides Us and Vs is congruent to 2 modulo n n divides V2js for some j < r. An extra strong Lucas pseudoprime is also a strong Lucas pseudoprime. and are both 1, the 324th Fibonacci number is a multiple of 323, and the 378th is a

Fibonacci pseudoprimes
A Fibonacci pseudoprime is a composite number n for which Vn is congruent to P modulo n when Q = 1. A strong Fibonacci pseudoprime may be defined as a composite number which is a Fibonacci pseudoprime for all P. It follows (see Mller and Oswald) that in this case: 1. An odd composite integer n is also a Carmichael number 2. 2(pi + 1) | (n 1) or 2(pi + 1) | (n pi) for every prime pi dividing n. The smallest example of a strong Fibonacci pseudoprime is 443372888629441, which has factors 17, 31, 41, 43, 89, 97, 167 and 331. It is conjectured that there are no even Fibonacci pseudoprimes (see Somer).

Fibonacci pseudoprime

21

References
Richard E. Crandall; Carl Pomerance (2001). Prime numbers: A computational approach. Springer-Verlag. pp.131132. ISBN0-387-94777-9. Hans Riesel (1994). Prime Numbers and Computer Methods for Factorization. Progress in Mathematics. 126 (2nd ed ed.). Birkhuser. p.130. ISBN0-8176-3743-5. Mller, Winfried B. and Alan Oswald. "Generalized Fibonacci Pseudoprimes and Probable Primes." In G.E. Bergum et al., eds. Applications of Fibonacci Numbers. Volume 5. Dordrecht: Kluwer, 1993. 459-464. Somer, Lawrence. "On Even Fibonacci Pseudoprimes." In G.E. Bergum et al., eds. Applications of Fibonacci Numbers. Volume 4. Dordrecht: Kluwer, 1991. 277-288. Richard K. Guy (2004). Unsolved Problems in Number Theory. Springer-Verlag. p.45. ISBN0-387-20860-7.

Anderson, Peter G. Fibonacci Pseudoprimes, their factors, and their entry points. [1] Anderson, Peter G. Fibonacci Pseudoprimes under 2,217,967,487 and their factors. [2] Weisstein, Eric W., "Fibonacci Pseudoprime [3]" from MathWorld. Weisstein, Eric W., "Lucas Pseudoprime [4]" from MathWorld.

Weisstein, Eric W., "Strong Lucas Pseudoprime [5]" from MathWorld. Weisstein, Eric W., "Extra Strong Lucas Pseudoprime [6]" from MathWorld.

References
[1] [2] [3] [4] [5] [6] http:/ / www. cs. rit. edu/ usr/ local/ pub/ pga/ fpp_and_entry_pts http:/ / www. cs. rit. edu/ usr/ local/ pub/ pga/ fibonacci_pp http:/ / mathworld. wolfram. com/ FibonacciPseudoprime. html http:/ / mathworld. wolfram. com/ LucasPseudoprime. html http:/ / mathworld. wolfram. com/ StrongLucasPseudoprime. html http:/ / mathworld. wolfram. com/ ExtraStrongLucasPseudoprime. html

Fibonacci retracement

22

Fibonacci retracement
Fibonacci retracements are a method of technical analysis for determining support and resistance levels. They are named after their use of the Fibonacci sequence. Fibonacci retracement is based on the idea that markets will retrace a predictable portion of a move, after which they will continue to move in the original direction. Fibonacci retracement is created by taking two extreme points on a chart and dividing the vertical distance by the key Fibonacci ratios. 0.0% is considered to be the start of the retracement, while 100.0% is a complete reversal to the original part of the move. Once these levels are identified, horizontal lines are drawn and used to identify possible support and resistance levels

Fibonacci retracement levels shown on the USD/CAD currency pair. In this case, price retraced approximately 38.2% of a move down before continuing.

Fibonacci ratios
Fibonacci ratios are mathematical relationships, expressed as ratios, derived from the Fibonacci sequence. The key Fibonacci ratios are 0%, 23.6%, 38.2%, 50%, 61.8% and 100%.

The key Fibonacci ratio of 0.618 is derived by dividing any number in the sequence by the number that immediately follows it. For example: 8/13 is approximately 0.6154, and 55/89 is approximately 0.6180.

The 0.382 ratio is found by dividing any number in the sequence by the number that is found two places to the right. For example: 34/89 is approximately 0.3820.

The 0.236 ratio is found by dividing any number in the sequence by the number that is three places to the right. For example: 55/233 is approximately 0.2361.

The 0 ratio is :

Fibonacci retracement

23

Other ratios
The 0.764 ratio is the result of subtracting 0.236 from the number 1.

## The 0.786 ratio is :

The 0.500 ratio is derived from dividing the number 1 (third number in the sequence) by the number 2 (forth number in the sequence).

Stevens, Leigh (2002). Essential technical analysis: tools and techniques to spot market trends. New York: Wiley. ISBN047115279X. OCLC48532501. Brown, Constance M. (2008). Fibonacci analysis. New York: Bloomberg Press. ISBN1576602613. Posamentier, Alfred S.; Lehmann, Ingmar (2007). The fabulous Fibonacci numbers. Amherst, NY: Prometheus Books. ISBN1591024757.

What is Fibonacci retracement, and where do the ratios that are used come from? [1] at investopedia.com Fibonacci Retracements [2] at stockcharts.com How to draw Fibonacci retracement, and how to analyze it? [3] at lollymotion.com

References
[1] http:/ / www. investopedia. com/ ask/ answers/ 05/ FibonacciRetracement. asp [2] http:/ / stockcharts. com/ school/ doku. php?id=chart_school:chart_analysis:fibonacci_retracemen [3] http:/ / lollymotion. com/ fibonacci/ fibonacci-retracement

24

## Fibonacci search technique

The Fibonacci search technique is a method of searching a sorted array using a divide and conquer algorithm that narrows down possible locations with the aid of Fibonacci numbers. Compared to binary search, Fibonacci search examines locations whose addresses have lower dispersion. Therefore, when the elements being searched have non-uniform access memory storage (i.e., the time needed to access a storage location varies depending on the location previously accessed), the Fibonacci search has an advantage over binary search in slightly reducing the average time needed to access a storage location. The typical example of non-uniform access storage is that of a magnetic tape, where the time to access a particular element is proportional to its distance from the element currently under the tape's head. Note, however, that large arrays not fitting in cache or even in RAM can also be considered as non-uniform access examples. Fibonacci search has a complexity of O(log(x)) (see Big O notation).

Algorithm

## Fibonacci search technique

25

Golden section search

References
David E. Ferguson, "Fibonaccian searching", Communications of the ACM, vol. 3 , is. 12, p.648, Dec. 1960. Manolis Lourakis, "Fibonaccian search in C". [1]. Retrieved January 18, 2007. Implements Ferguson's algorithm. Donald E. Knuth, "The Art of Computer Programming (second edition)", vol. 3 , p.418, Nov. 2003.

References
[1] http:/ / www. ics. forth. gr/ ~lourakis/ fibsrch/

Fibonacci triangle
Hosoya's triangle or the Fibonacci triangle is a triangular arrangement of numbers (like Pascal's triangle) based on the Fibonacci numbers. Each number is the sum of the two numbers above in either the left diagonal or the right diagonal. The first few rows are: 1 1 2 3 5 8 13 21 34 55 34 21 42 13 26 39 8 16 24 40 5 10 15 25 40 3 6 9 15 24 39 2 4 6 10 16 26 42 1 2 3 5 8 13 21 34 1 2 3 5 8 13 21 34 55

(See (sequence A058071 [1] in OEIS)). The recurrence relation is H(0,0) =H(1,0) =H(1,1) =H(2,1)=1 and H(n,j) =H(n1,j)+H(n2,j) or H(n,j) =H(n1,j1)+H(n2,j2). The entries in the triangle satisfy the identity H(n,i)=F(i+1)F(ni+1). Thus, the two outermost diagonals are the Fibonacci numbers, while the numbers on the middle vertical line are the squares of the Fibonacci numbers. All the other numbers in the triangle are the product of two distinct Fibonacci numbers greater than 1. The row sums are the convolved Fibonacci numbers (A001629 [2]).

Fibonacci triangle

26

References
Haruo Hosoya, "Fibonacci Triangle" The Fibonacci Quarterly 14 2 (1976): 173178 Thomas Koshy, Fibonacci and Lucas Numbers and Applications. New York: Wiley & Sons (2001): 187195 ramnatthan.ala(cit)

References
[1] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa058071 [2] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa001629

FibonacciSylvester expansion
In mathematics, the greedy algorithm for Egyptian fractions is a greedy algorithm, first described by Fibonacci, for transforming rational numbers into Egyptian fractions. An Egyptian fraction is a representation of an irreducible fraction as a sum of unit fractions, as e.g. 5/6 = 1/2 + 1/3. As the name indicates, these representations have been used as long ago as ancient Egypt, but the first published systematic method for constructing such expansions is described in the Liber Abaci (1202) of Leonardo of Pisa (Fibonacci). It is called a greedy algorithm because at each step the algorithm chooses greedily the largest possible unit fraction that can be used in any representation of the remaining fraction. Fibonacci actually lists several different methods for constructing Egyptian fraction representations (Sigler 2002, chapter II.7). He includes the greedy method as a last resort for situations when several simpler methods fail; see Egyptian fraction for a more detailed listing of these methods. As Salzer (1948) details, the greedy method, and extensions of it for the approximation of irrational numbers, have been rediscovered several times by modern mathematicians, earliest and most notably by Sylvester (1880); see for instance Cahen (1891) and Spiess (1907). A closely related expansion method that produces closer approximations at each step by allowing some unit fractions in the sum to be negative dates back to Lambert (1770). The expansion produced by this method for a number x is called the greedy Egyptian expansion, Sylvester expansion, or FibonacciSylvester expansion of x. However, the term Fibonacci expansion usually refers, not to this method, but to representation of integers as sums of Fibonacci numbers.

## Algorithm and examples

Fibonacci's algorithm expands the fraction x/y to be represented, by repeatedly performing the replacement

## (simplifying the second term in this replacement as necessary). For instance:

in this expansion, the denominator 3 of the first unit fraction is the result of rounding 15/7 up to the next larger integer, and the remaining fraction 2/15 is the result of simplifying (-15 mod 7)/153 = 6/45. The denominator of the second unit fraction, 8, is the result of rounding 15/2 up to the next larger integer, and the remaining fraction 1/120 is what is left from 7/15 after subtracting both 1/3 and 1/8. As each expansion step reduces the numerator of the remaining fraction to be expanded, this method always terminates with a finite expansion; however, compared to ancient Egyptian expansions or to more modern methods, this method may produce expansions that are quite long, with large denominators. For instance, this method expands

FibonacciSylvester expansion

27

## while other methods lead to the much better expansion

Wagon (1991) suggests an even more badly-behaved example, 31/311. The greedy method leads to an expansion with ten terms, the last of which has over 500 digits in its denominator; however, 31/311 has a much shorter non-greedy representation, 1/12 + 1/63 + 1/2799 + 1/8708.

## Sylvester's sequence and closest approximation

Sylvester's sequence 2, 3, 7, 43, 1807, ... can be viewed as generated by an infinite greedy expansion of this type for the number one, where at each step we choose the denominator instead of . Truncating this sequence to k terms and forming the corresponding Egyptian fraction, e.g. (for k = 4)

results in the closest possible underestimate of 1 by any k-term Egyptian fraction (Curtiss 1922; Soundararajan 2005). That is, for example, any Egyptian fraction for a number in the open interval (1805/1806,1) requires at least five terms. Curtiss (1922) describes an application of these closest-approximation results in lower-bounding the number of divisors of a perfect number, while Stong (1983) describes applications in group theory.

## Maximum-length expansions and congruence conditions

Any fraction x/y requires at most x terms in its greedy expansion. Mays (1987) and Freitag and Phillips (1999) examine the conditions under which x/y leads to an expansion with exactly x terms; these can be described in terms of congruence conditions on y. Every fraction 1/y requires one term in its expansion; the simplest such fraction is 1/1. Every fraction 2/y for odd y > 1 requires two terms in its expansion; the simplest such fraction is 2/3. A fraction 3/y requires three terms in its expansion if and only if y 1 (mod 6), for then -y mod x = 2 and y(y+2)/3 is odd, so the fraction remaining after a single step of the greedy expansion,

is in simplest terms. The simplest fraction 3/y with a three-term expansion is 3/7. A fraction 4/y requires four terms in its expansion if and only if y 1 or 17 (mod 24), for then the numerator -y mod x of the remaining fraction is 3 and the denominator is 1 (mod 6). The simplest fraction 4/y with a four-term expansion is 4/17. The ErdsStraus conjecture states that all fractions 4/y have an expansion with three or fewer terms, but when y 1 or 17 (mod 24) such expansions must be found by methods other than the greedy algorithm. More generally the sequence of the smallest denominators y leading to the longest expansion for each x is 1, 3, 7, 17, 31, 109, 253, 97, 271, ... (sequence A048860 [1] in OEIS).

## Approximation of polynomial roots

Stratemeyer (1930) and Salzer (1947) describe a method of finding an accurate approximation for the roots of a polynomial based on the greedy method. Their algorithm computes the greedy expansion of a root; at each step in this expansion it maintains an auxiliary polynomial that has as its root the remaining fraction to be expanded. Consider as an example applying this method to find the greedy expansion of the golden ratio, one of the two solutions of the polynomial equation P0(x) = x2 - x - 1 = 0. The algorithm of Stratemeyer and Salzer performs the

FibonacciSylvester expansion following sequence of steps: 1. Since P0(x) < 0 for x = 1, and P0(x) > 0 for all x 2, there must be a root of P0(x) between 1 and 2. That is, the first term of the greedy expansion of the golden ratio is 1/1. If x1 is the remaining fraction after the first step of the greedy expansion, it satisfies the equation P0(x1 + 1) = 0, which can be expanded as P1(x1) = x12 + x1 - 1 = 0. 2. Since P1(x) < 0 for x = 1/2, and P1(x) > 0 for all x > 1, the root of P1 lies between 1/2 and 1, and the first term in its greedy expansion (the second term in the greedy expansion for the golden ratio) is 1/2. If x2 is the remaining fraction after this step of the greedy expansion, it satisfies the equation P1(x2 + 1/2) = 0, which can be expanded as P2(x2) = 4x22 + 8x2 - 1 = 0. 3. Since P2(x) < 0 for x = 1/9, and P2(x) > 0 for all x > 1/8, the next term in the greedy expansion is 1/9. If x3 is the remaining fraction after this step of the greedy expansion, it satisfies the equation P2(x3 + 1/9) = 0, which can again be expanded as a polynomial equation with integer coefficients, P3(x3) = 324x32 + 720x3 - 5 = 0. Continuing this approximation process eventually produces the greedy expansion for the golden ratio, (sequence A117116 [2] in OEIS).

28

## Other integer sequences

The length, minimum denominator, and maximum denominator of the greedy expansion for all fractions with small numerators and denominators can be found in the On-Line Encyclopedia of Integer Sequences as sequences A050205 [3], A050206 [4], and A050210 [5], respectively. In addition, the greedy expansion of any irrational number leads to an infinite increasing sequence of integers, and the OEIS contains expansions of several well known constants [6]. Some additional entries in the OEIS [7], though not labeled as being produced by the greedy algorithm, appear to be of the same type.

Related expansions
In general, if one wants an Egyptian fraction expansion in which the denominators are constrained in some way, it is possible to define a greedy algorithm in which at each step one chooses the expansion

where d is chosen, among all possible values satisfying the constraints, as small as possible such that xd > y and such that d is distinct from all previously chosen denominators. For instance, the Engel expansion can be viewed as an algorithm of this type in which each successive denominator must be a multiple of the previous one. However, it may be difficult to determine whether an algorithm of this type can always succeed in finding a finite expansion. In particular, the odd greedy expansion of a fraction x/y is formed by a greedy algorithm of this type in which all denominators are constrained to be odd numbers; it is known that, whenever y is odd, there is a finite Egyptian fraction expansion in which all denominators are odd, but it is not known whether the odd greedy expansion is always finite.

FibonacciSylvester expansion

29

References
Cahen, E. (1891). "Note sur un dveloppement des quantits numriques, qui presente quelque analogie avec celui en fractions continues". Nouvelles Annales des Mathmatiques, Ser. 3 10: 508514. Curtiss, D. R. (1922). "On Kellogg's diophantine problem" [8]. American Mathematical Monthly 29 (10): 380387. doi:10.2307/2299023. Freitag, H. T.; Phillips, G. M. (1999). "Sylvester's algorithm and Fibonacci numbers". Applications of Fibonacci numbers, Vol. 8 (Rochester, NY, 1998). Dordrecht: Kluwer Acad. Publ.. pp.155163. MR1737669. Lambert, J. H. (1770). Beytrge zum Gebrauche der Mathematik und deren Anwendung. Berlin: Zweyter Theil. pp.99104. Mays, Michael (1987). "A worst case of the Fibonacci-Sylvester expansion". Journal of Combinatorial Mathematics and Combinatorial Computing 1: 141148. MR0888838. Salzer, H. E. (1947). "The approximation of numbers as sums of reciprocals" [9]. American Mathematical Monthly 54 (3): 135142. doi:10.2307/2305906. MR0020339. Salzer, H. E. (1948). "Further remarks on the approximation of numbers as sums of reciprocals" [10]. American Mathematical Monthly 55 (6): 350356. doi:10.2307/2304960. MR0025512. Sigler, Laurence E. (trans.) (2002). Fibonacci's Liber Abaci. Springer-Verlag. ISBN0-387-95419-8. Soundararajan, K. (2005). Approximating 1 from below using n Egyptian fractions. arXiv:math.CA/0502247. Spiess, O. (1907). "ber eine Klasse unendlicher Reihen". Archiv der Mathematik und Physik, Ser. 3 12: 124134. Stong, R. E. (1983). "Pseudofree actions and the greedy algorithm". Mathematische Annalen 265 (4): 501512. doi:10.1007/BF01455950. MR0721884. Stratemeyer, G. (1930). "Stammbruchentwickelungen fr die Quadratwurzel aus einer rationalen Zahl". Mathematische Zeitschrift 31: 767768. doi:10.1007/BF01246446. Sylvester, J. J. (1880). "On a point in the theory of vulgar fractions" [11]. American Journal of Mathematics 3 (4): 332335. doi:10.2307/2369261. Wagon, S. (1991). Mathematica in Action. W. H. Freeman. pp.271277.

References
[1] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa048860 [2] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa117116 [3] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa050205 [4] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa050206 [5] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa050210 [6] http:/ / www. research. att. com/ ~njas/ sequences/ ?q=greedy-Egyptian-fraction-expansion [7] http:/ / www. research. att. com/ ~njas/ sequences/ ?q=Egyptian-fraction-for [8] http:/ / jstor. org/ stable/ 2299023 [9] http:/ / jstor. org/ stable/ 2305906 [10] http:/ / jstor. org/ stable/ 2304960 [11] http:/ / jstor. org/ stable/ 2369261

30

## Lagged Fibonacci generator

A Lagged Fibonacci generator (LFG) is an example of a pseudorandom number generator. This class of random number generator is aimed at being an improvement on the 'standard' linear congruential generator. These are based on a generalisation of the Fibonacci sequence. The Fibonacci sequence may be described by the recurrence relation:

Hence, the new term is the sum of the last two terms in the sequence. This can be generalised to the sequence: In which case, the new term is some combination of any two previous terms. m is usually a power of 2 (m = 2M), often 232 or 264. The operator denotes a general binary operation. This may be either addition, subtraction, multiplication, or the bitwise arithmetic exclusive-or operator (XOR). The theory of this type of generator is rather complex, and it may not be sufficient simply to choose random values for j and k. These generators also tend to be very sensitive to initialisation. Generators of this type employ k words of state (they 'remember' the last k values). If the operation used is addition, then the generator is described as an Additive Lagged Fibonacci Generator or ALFG, if multiplication is used, it is a Multiplicative Lagged Fibonacci Generator or MLFG, and if the XOR operation is used, it is called a Two-tap generalised feedback shift register or GFSR. The Mersenne twister algorithm is a variation on a GFSR. The GFSR is also related to the Linear Feedback Shift Register, or LFSR.

## Properties of lagged Fibonacci generators

Lagged Fibonacci generators have a maximum period of (2k - 1)*2M-1 if addition or subtraction is used, and (2k-1)*k if exclusive-or operations are used to combine the previous values. If, on the other hand, multiplication is used, the maximum period is (2k - 1)*2M-3, or 1/4 of period of the additive case. For the generator to achieve this maximum period, the polynomial: y = xk + xj + 1 must be primitive over the integers mod 2. Values of j and k satisfying this constraint have been published in the literature. Popular pairs are: {j = 7, k = 10}, {j = 5, k = 17}, {j = 24, k = 55}, {j = 65, k = 71}, {j = 128, k = 159} [1], {j = 6, k = 31}, {j = 31, k = 63}, {j = 97, k = 127}, {j = 353, k = 521}, {j = 168, k = 521}, {j = 334, k = 607}, {j = 273, k = 607}, {j = 418, k = 1279} [2] Another list of possible values for j and k is on page 29 of volume 2 of The Art of Computer Programming: (24,55), (38,89), (37,100), (30,127), (83,258), (107,378), (273,607), (1029,2281), (576,3217), (4187,9689), (7083,19937), (9739,23209) Note that the smaller number have short periods (only a few "random" numbers are generated before the first "random" number is repeated and the sequence restarts). It is required that at least one of the first k values chosen to initialise the generator be odd. It has been suggested that good ratios between j and k are approximately the golden ratio[3] .

31

## Problems with LFGs

In a paper on four-tap shift registers, Robert M. Ziff states that "It is now widely known that such generators, in particular with the two-tap rules such as R(103, 250), have serious deficiencies. Marsaglia observed very poor behavior with R(24,55) and smaller generators, and advised against using generators of this type altogether. ... The basic problem of two-tap generators R(a, b) is that they have a built-in three-point correlation between , , and , simply given by the generator itself ... While these correlations are spread over the size of the generator itself, they can evidently still lead to significant errors."[4] . The initialization of LFGs is a very complex problem. The output of LFGs is very sensitive to initial conditions, and statistical defects may appear initially but also periodically in the output sequence unless extreme care is taken . Another potential problem with LFGs is that the mathematical theory behind them is incomplete, making it necessary to rely on statistical tests rather than theoretical performance.

Usage
Freeciv uses a lagged Fibonacci generator with {j = 24, k = 55} for its random number generator. The Boost library includes an implementation of a lagged Fibonacci generator. MATLAB uses a {j=??, k = 32} generator for its rand() function. The Oracle Database implements this generator in its DBMS_RANDOM package (available in Oracle 8 and newer versions). .NET CLR uses a lagged Fibonacci generator for its System.Random generator (http://msdn.microsoft.com/ en-us/library/system.random.aspx).

References
[1] http:/ / www. ccs. uky. edu/ csep/ RN/ RN. html [2] http:/ / www. nersc. gov/ nusers/ resources/ software/ libs/ math/ random/ www2. 0/ DOCS/ www/ parameters. html [3] "Uniform random number generators for supercomputers", Richard Brent, Proc. of Fifth Australian Supercomputer Conference, Melbourne, Dec. 1992, pp. 704-706 [4] "Four-tap shift-register-sequence random-number generators" (http:/ / arxiv. org/ abs/ cond-mat/ 9710104:), Robert M. Ziff, Computers in Physics, 12(4), Jul/Aug 1998, pp. 385392

Negafibonacci

32

Negafibonacci
Numeral systems by culture Hindu-Arabic numerals Western Arabic Eastern Arabic Indian family Burmese Khmer Mongolian Thai

East Asian numerals Chinese Japanese Suzhou Korean Vietnamese Counting rods Alphabetic numerals Abjad Armenian ryabhaa Cyrillic Ge'ez Greek (Ionian) Hebrew

Other systems Aegean Attic Babylonian Brahmi Egyptian Etruscan Inuit Mayan Quipu Roman Sumerian Urnfield

List of numeral system topics Positional systems by base Decimal (10) 1, 2, 3, 4, 5, 6, 8, 12, 16, 20, 30, 36, 60 more

Definition
In mathematics, negaFibonacci numbers are the negatively indexed elements of the Fibonacci sequence. The negaFibonacci numbers are defined inductively by the recurrence relation: F-1=1, F-2=-1, Fn=F(n+2)F(n+1). They may be defined by the formula Fn=(1)n+1Fn. The first 10 negaFibonacci numbers are

Negafibonacci

33

## n F(n) -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 1 -1 2 -3 5 -8 13 -21 34 -55

Integer representation
Any integer can be uniquely represented[1] as a sum of negaFibonacci numbers in which no two consecutive negaFibonacci numbers are used. For example: -11 = F-4 + F-6 = (-3) + (-8) 12 = F-2 + F-7 = (-1) + 13 24 = F-1 + F-4 + F-6 + F-9 = 1 + (-3) + (-8) + 34 -43 = F-2 + F-7 + F-10 = (-1) + 13 + (-55) 0 is represented by the empty sum.

Note that 0 = F-1 + F-2, for example, so the uniqueness of the representation does depend on the condition that no two consecutive negafibonacci numbers are used. This gives a system of coding integers, similar to the representation of Zeckendorf's theorem for coding numbers using a binary representation. In the string representing the integer x, the nth digit is 1 if Fn appears in the sum that represents x; that digit is 0 otherwise. For example, 24 may be represented by the string 100101001, which has the digit 1 in places 9, 6, 4, and 1, because 24 = F-1 + F-4 + F-6 + F-9. The integer x is represented by a string of odd length if and only if .

Identities
Relationship to the normal, positive Fibonacci number sequence:

Notes
[1] Knuth, Donald. "Negafibonacci Numbers and the Hyperbolic Plane" Paper presented at the annual meeting of the Mathematical Association of America, The Fairmont Hotel, San Jose, CA. 2008-12-11 <http://www.allacademic.com/meta/p206842_index.html>

NegaFibonacci coding

34

NegaFibonacci coding
Numeral systems by culture Hindu-Arabic numerals Western Arabic Eastern Arabic Indian family Burmese Khmer Mongolian Thai

East Asian numerals Chinese Japanese Suzhou Korean Vietnamese Counting rods Alphabetic numerals Abjad Armenian ryabhaa Cyrillic Ge'ez Greek (Ionian) Hebrew

Other systems Aegean Attic Babylonian Brahmi Egyptian Etruscan Inuit Mayan Quipu Roman Sumerian Urnfield

List of numeral system topics Positional systems by base Decimal (10) 1, 2, 3, 4, 5, 6, 8, 12, 16, 20, 30, 36, 60 more

In mathematics, negaFibonacci coding is a universal code which encodes integers into binary code words. It is similar to Fibonacci coding, except that it allows both positive and negative integers to be represented. All codes end with "11" and have no "11" before the end. The code for the integers from -11 to 11 is given below. xx negaFibonacci representation -11 101000 -10 101001 -9 100010 -8 100000 -7 100001 -6 100100 -5 100101 -4 1010 -3 1000 -2 1001 -1 10 0 0 1 1 negaFibonacci code 0001011 1001011 0100011 0000011 1000011 0010011 1010011 01011 00011 10011 011 01 11

NegaFibonacci coding 2 3 4 5 6 7 8 9 10 11 100 101 10010 10000 10001 10100 10101 1001010 1001000 1001001 0011 1011 010011 000011 100011 001011 101011 01010011 00010011 10010011

35

The Fibonacci code is closely related to negaFibonacci representation, a positional numeral system sometimes used by mathematicians. The negaFibonacci code for a particular integer is exactly that of the integer's negaFibonacci representation, except with the order of its digits reversed and an additional "1" appended to the end. The negaFibonacci code for all negative numbers has an odd number of digits, while those of all positive numbers have an even number of digits. To encode an integer X: 1. Find the largest negafibonacci number with a value equal to or less than X; subtract this number from X, keeping track of the remainder. 2. If the number we subtracted was the Nth unique negaFibonacci number, put a one in the Nth digit of our output. 3. Repeat the previous steps, substituting our remainder for X, until we reach a remainder of 0. 4. Place a one after the last naturally occurring one in our output. To decode a token in the code, remove the last "1", assign the remaining bits the values -1,2,-3,5,-8,13... (the negafibonacci numbers), and add the "1" bits.

Pisano period

36

Pisano period
In number theory, the nth Pisano period, written (n), is the period with which the sequence of Fibonacci numbers, modulo n repeats. For example, the Fibonacci numbers mod 3 are 0, 1, 1, 2, 0, 2, 2, 1, 0, 1, 1, etc., with the first eight numbers repeating, so (3) = 8. Pisano periods are named after Leonardo Pisano, better known as Fibonacci.

Tables
The first Pisano periods (sequence A001175 readability) are:
n (n) nr. of 0s 1 2 3 4 5 6 7 8 9 1 3 8 6 20 24 16 12 24 1 1 2 1 4 2 2 2 2 4 0 011 0112 0221 011231 01123 03314 04432 02241 011235213415 055431453251 01123516 06654261 011235 055271 011235843718 088764156281 011235831459437 077415617853819 099875279651673 033695493257291 [1]

in OEIS) and their cycles (with spaces before the zeros for

cycle

10 60

Onward the Pisano periods are 10, 24, 28, 48, 40, 24, 36, 24, 18, 60, 16, 30, 48, 24, 100 ... For n > 2 the period is even, because alternatingly F(n)2 is one more and one less than F(n1)F(n+1) (Cassini's identity). The period is relatively small, 4k + 2, for n = F (2k) + F (2k + 2), i.e. Lucas number L (2k + 1), with k a positive integer. This is because F(2k1) = F (2k+1) and F(2k) = F (2k), and the latter is congruent to F(2k+2)modulon, showing that the period is a divisor of 4k+2; the period cannot be 2k+1 or less because the first 2k+1 Fibonacci numbers from 0 are less than n.
k 1 4 2 11 3 29 4 76 5 199 6 521 n (n) 6 10 14 18 22 26 first half of cycle (with selected second halfs) 0, 1, 1, 2 (, 3, 1) 0, 1, 1, 2, 3, 5 (, 8, 2, 10, 1) 0, 1, 1, 2, 3, 5, 8, 13 (, 21, 5, 26, 2, 28, 1) 0, 1, 1, 2, 3, 5, 8, 13, 21, 34 (, 55, 13, 68, 5, 73, 2, 75, 1) 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89 (, 144, 34, 178, 13, 191, 5, 196, 2, 198, 1) 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1597 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1597, 2584, 4181

## 7 1364 30 8 3571 34 9 9349 38

Pisano period The second half of the cycle, which is of course equal to the part on the left of 0, consists of alternatingly numbers F(2m+1) and nF(2m), with m decreasing. Furthermore, the period is 4k for n = F(2k), and 8k+4 for n = F(2k+1). The number of occurrences of 0 per cycle is 1, 2, or 4. Let p be the number after the first 0 after the combination 0, 1. Let the distance between the 0s be q. There is one 0 in a cycle, obviously, if p=1. This is only possible if q is even or n is 1 or 2. Otherwise there are two 0s in a cycle if p21. This is only possible if q is even. Otherwise there are four 0s in a cycle. This is the case if q is odd and n is not 1 or 2. For generalized Fibonacci sequences (satisfying the same recurrence relation, but with other initial values, e.g. the Lucas numbers) the number of occurrences of 0 per cycle is 0, 1, 2, or 4. Also, it can be proven that (n) 6n, with equality if and only if n=25k, for k1, the first examples being (10)=60 and (50)=300.

37

Number theory
Pisano periods can be analyzed using algebraic number theory. If m and n are coprime, then example, and so by the Chinese remainder theorem: two numbers Thus it

are congruent mod mn if and only if they are congruent mod m and mod n, assuming these latter are coprime. For suffices to compute Pisano periods for prime powers For prime numbers p, these can be analyzed by using Binet's formula: where is the golden ratio

), then

and

## can be expressed as integers mod p,

and thus Binet's formula can be expressed over integers mod p, and thus the Pisano period divides the totient , since any power (such as ) has period dividing as this is the order of the group of units mod p. This first occurs for n=11, where 42=165(mod11) and 26=121(mod11) and 43=121(mod11) so 4=5, 6=1/2 and 1/5=3, yielding =(1+4)6 =308(mod11) and the formula

Another example, which shows that the period can properly divide p1, is (29)=14. If 5 is not a quadratic residue (and p2,5), then Binet's formula is instead defined over the quadratic extension field (Z/p)[5], which has p2 elements and whose group of units thus has order p21, and thus the Pisano period divides p21. For example, for p=3 one has (3)=8 which equals 321=8; for p=7, one has (7)=16, which properly divides 721=48. This analysis fails for p=2 and p=5 since in these cases 2 and 5 are zero divisors, so one must be careful in interpreting 1/2 or5. For p=2, 5 is congruent to 1 mod2, but the Pisano period is not p1=1, but rather3. For p=5, the Pisano period is (5)=20=5(51), which does not divide p1=4 or p21=24.

Pisano period

38

Sums
Using F(0) + F(1) + F(2) + + F(k) = F(k + 2) 1, it follows that the sum of (n) consecutive Fibonacci numbers is a multiple of n. Moreover, for the values in the table the sum of (n) consecutive Fibonacci numbers is n times the ((n)/2 + 1)th element:

Powers of 10
The Pisano periods when n is a power of 10 are 60, 300, 1500, 15000, 150000, ... (sequence A096363 Dov Jarden proved that for n greater than 2 the periodicity mod10n is 1510n1.[3]
[2]

in OEIS).

Cultural references
The Fibonacci sequence mod 5 (Pisano period 20, with 4 zeros) is featured in the episode "The Case of the Willing Parrot" of the TV Mathnet, where the sequence is depicted as tiles on a wall.

## Fibonacci integer sequences modulo n

One can consider Fibonacci integer sequences and take them modulo n, or put differently, consider Fibonacci sequences in the ring Z/n. The period is a divisor of (n). The number of occurrences of 0 per cycle is 0, 1, 2, or 4. If n is not a prime the cycles include those that are multiples of the cycles for the divisors. For example, for n = 10 the extra cycles include those for n = 2 multiplied by 5, and for n = 5 multiplied by 2. Table of the extra cycles:
n 1 2 3 4 5 6 7 8 9 0 0 0, 022 0 0, 0224 0442, 033 0 0, 022462, 044, 066426 0, 0336 0663 02246325 05531452, 03362134 04415643 033617 077653, 134732574372, 145167541563 022461786527 077538213472, 044832573145 055167426854 033213 1342 multiples other cycles

## 10 0, 02246 06628 08864 04482, 055, 2684 134718976392

Pisano period

39

References
Johnson, R. C. (2008). "Fibonacci numbers and matrices" [4]. Engstrom, H. T. (1931). "On sequences defined by linear recurrence relations". Trans. Am. Math. Soc. 33 (1): 210218. doi:10.1090/S0002-9947-1931-1501585-5. JSTOR1989467 MR1501585 Ward, Morgan (1931). "The characteristic number of a sequence of integers satisfying a linear recursion relation". Trans. Am. Math. Soc. 33 (1): 153165. doi:10.1090/S0002-9947-1931-1501582-X. JSTOR1989464 Ward, Morgan (1933). "The arithmetical theory of linear recurring series". Trans. Am. Math. Soc. 35 (3): 600628. doi:10.1090/S0002-9947-1933-1501705-4. JSTOR1989851 Zierler, Neal (1959). "Linear recurring sequences". J. SIAM 7 (1): 3138. doi:10.1137/0107003. JSTOR2099002 MR0101979 Laxton, R. R. (1969). "On groups of linear recurrences". Duke Math J. 36 (4): 721736. doi:10.1215/S0012-7094-69-03687-4. MR0258781 Brent, Richard P. (1994). "On the periods of generalized Fibonacci sequences". Mathematics of Computation 63 (207): 389401. Bibcode:1994MaCom..63..389B. JSTOR2153583 MR1216256
[1] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa001175 [2] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa096363 [3] Weisstein, Eric W., " Pisano Period (http:/ / mathworld. wolfram. com/ PisanoPeriod. html)" from MathWorld. [4] http:/ / maths. dur. ac. uk/ ~dma0rcj/ PED/ fib. pdf

Renault, N. S.. "Table of Pisano periods" (http://www.math.temple.edu/~renault/fibonacci/fiblist.html).

## Reciprocal Fibonacci constant

The reciprocal Fibonacci constant, or , is defined as the sum of the reciprocals of the Fibonacci numbers:

The ratio of successive terms in this sum tends to the reciprocal of the golden ratio. Since this is less than 1, the ratio test shows that the sum converges. The value of is known to be approximately
[1]

No closed formula for is known, but Gosper describes an algorithm for fast numerical approximation of its value. The reciprocal Fibonacci series itself provides O(k) digits of accuracy for k terms of expansion, while Gosper's accelerated series provides O(k2) digits. [2] is known to be irrational; this property was conjectured by Paul Erds, Ronald Graham, and Leonard Carlitz, and proved in 1989 by Richard Andr-Jeannin.[3] The continued fraction representation of the constant is:
[4]

## Reciprocal Fibonacci constant

40

References
[1] (sequence A079586 (http:/ / en. wikipedia. org/ wiki/ Oeis:a079586) in OEIS) [2] Gosper, William R. (1974), Acceleration of Series (http:/ / dspace. mit. edu/ handle/ 1721. 1/ 6088), Artificial Intelligence Memo #304, Artificial Intelligence Laboratory, Massachusetts Institute of Technology, pp.p.66, . [3] Andr-Jeannin, Richard (1989), "Irrationalit de la somme des inverses de certaines suites rcurrentes", C. R. Acad. Sci. Paris Sr. I Math. 308 (19): 539541, MR0999451 [4] (sequence A079587 (http:/ / en. wikipedia. org/ wiki/ Oeis:a079587) in OEIS)

Weisstein, Eric W., " Reciprocal Fibonacci Constant (http://mathworld.wolfram.com/ ReciprocalFibonacciConstant.html)" from MathWorld.

YoungFibonacci lattice
In mathematics, the YoungFibonacci graph and YoungFibonacci lattice, named after Alfred Young and Leonardo Fibonacci, are two closely related structures involving sequences of the digits 1 and 2. Any digit sequence of this type can be assigned a rank, the sum of its digits: for instance, the rank of 11212 is 1+1+2+1+2=7. As was already known in ancient India, the number of sequences with a given rank is a Fibonacci number. The YoungFibonacci lattice is an infinite modular lattice having these digit sequences as its elements, compatible with this rank structure. The YoungFibonacci graph is the graph of this lattice, and has a vertex for each digit sequence.

## The YoungFibonacci graph, the Hasse diagram of the YoungFibonacci lattice.

The YoungFibonacci graph and the YoungFibonacci lattice were both initially studied in two papers by Fomin (1988) and Stanley (1988). They are named after the closely related Young's lattice and after the Fibonacci number of their elements at any given rank.

## Digit sequences with a given rank

A digit sequence with rank r may be formed either by adding the digit 2 to a sequence with rank r2, or by adding the digit 1 to a sequence with rank r1. If (r) is the function that maps r to the number of different digit sequences of that rank, therefore, satisfies the recurrence relation (r)=(r2)+(r1) defining the Fibonacci numbers, but with slightly different initial conditions: (0)=(1)=1 (there is one rank-0 string, the empty string, and one rank-1 string, consisting of the single digit 1. These initial conditions cause the sequence of values of to be shifted by one position from the Fibonacci numbers: (r)=Fr+1 where Fi denotes the ith Fibonacci number. In the ancient Indian study of prosody, the Fibonacci numbers were used to count the number of different sequences of short and long syllables with a given total length; if the digit 1 corresponds to a short syllable, and the digit 2 corresponds to a long syllable, the rank of a digit sequence measures the total length of the corresponding sequence

YoungFibonacci lattice of syllables. See the Fibonacci number article for details.

41

## Graphs of digit sequences

The YoungFibonacci graph is an infinite graph, with a vertex for each string of the digits 1 and 2 (including the empty string). The neighbors of a string s are the strings formed from s by one the following operations: 1. 2. 3. 4. Insert a 1 into s, prior to the leftmost 1 (or anywhere in s if it does not already contain a 1). Change the leftmost 1 of s into a 2. Remove the leftmost 1 from s. Change a 2 that does not have a 1 to the left of it into a 1.

It is straightforward to verify that each operation can be inverted: operations 1 and 3 are inverse to each other, as are operations 2 and 4. Therefore, the resulting graph may be considered to be undirected. However, it is usually considered to be a directed acyclic graph in which each edge connects from a vertex of lower rank to a vertex of higher rank. As both Fomin (1988) and Stanley (1988) observe, this graph has the following properties: It is connected: any nonempty string may have its rank reduced by some operation, so there is a sequence of operations leading from it to the empty string, reversing which gives a directed path in the graph from the empty string to every other vertex. It is compatible with the rank structure: every directed path has length equal to the difference in ranks of its endpoints. For every two nodes u and v, the number of common immediate predecessors of u and v equals the number of common immediate successors of u and v; this number is either zero or one. The out-degree of every vertex equals one plus its in-degree. Fomin (1988) calls a graph with these properties a Y-graph; Stanley (1988) calls a graph with a weaker version of these properties (in which the numbers of common predecessors and common successors of any pair of nodes must be equal but may be greater than one) the graph of a differential poset.

## Partial order and lattice structure

The transitive closure of the YoungFibonacci graph is a partial order. As Stanley (1988) shows, any two vertices x and y have a unique greatest common predecessor in this order (their meet) and a unique least common successor (their join); thus, this order is a lattice, called the YoungFibonacci lattice. To find the meet of x and y, one may first test whether one of x and y is a predecessor of the other. A string x is a predecessor of another string y in this order exactly when the number of 2 digits remaining in y, after removing the longest common suffix of x and y, is at least as large as the number of digits remaining in x after removing the common suffix. In this case, the meet is whichever of x and y is the predecessor. In a second case, if neither x nor y is the predecessor of the other, but one or both of them begins with a 1 digit, the meet is unchanged if these initial digits are removed. And finally, if both x and y begin with the digit 2, the meet of x and y may be found by removing this digit from both of them, finding the meet of the resulting suffixes, and adding the 2 back to the start. A common successor of x and y (though not necessarily the least common successor) may be found by taking a string of 2 digits with length equal to the longer of x and y. The least common successor is then the meet of the finitely many strings that are common successors of x and y and predecessors of this string of 2s. As Stanley (1988) further observes, the YoungFibonacci lattice is modular. Fomin (1988) incorrectly claims that it is distributive; the sublattice formed by the strings {21,22,121,211,221} forms a diamond sublattice, forbidden in distributive lattices.

YoungFibonacci lattice

42

References
Fomin, S. V. (1988), "Generalized Robinson-Schensted-Knuth correspondence", Journal of Mathematical Sciences 41 (2): 979991, doi:10.1007/BF01247093. Translated from Zapiski Nauchnykh Seminarov Leningradskogo Otdeleniya Matematicheskogo Instituta im. V. A. Steklova AN SSSR 155: 156175, 1986. Stanley, Richard P. (1988), "Differential posets" [1], Journal of the American Mathematical Society 1 (4): 919961, doi:10.2307/1990995.

References
[1] http:/ / jstor. org/ stable/ 1990995

## The Fibonacci Association

The Fibonacci Association is a mathematical organization that specializes in the Fibonacci number sequence and a wide variety of related subjects, generalizations, and applications, including recurrence relations, combinatorial identities, binomial coefficients, prime numbers, pseudoprimes, continued fractions, the golden ratio, linear algebra, geometry, real analysis, and complex analysis. The organization was founded in 1963 by Verner E. Hoggatt Jr. of San Jose State College (now San Jose State University) and Brother Alfred Brousseau, F.S.C. of St. Mary's College (Moraga, California). Since the year of its founding, the Fibonacci Association has published an international mathematical journal, The Fibonacci Quarterly [1]. The Fibonacci Association also publishes Proceedings for its international conferences, held every two years since 1984. The 2008 conference, formally entitled the Thirteenth International Conference on Fibonacci Numbers and Their Applications, took place at the University of Patras (Greece), preceded by conferences at San Francisco State University (USA, 2006), Technische Universitt Braunschweig (Germany, 2004), Northern Arizona University (USA, 2002), and Institut Suprieur de Technologie (Luxemburg, 2000). The 2010 Conference was held at the Instituto de Matemticas de la UNAM, Morelia, Mexico, as announced at the Fibonacci Association website: [2]. An announcement for the 2012 Conference is forthcoming. Details regarding the early history of The Fibonacci Association are given Marjorie Bicknell-Johnson's "A Short History of The Fibonacci Quarterly", published in The Fibonacci Quarterly 25:1 (February 1987) 2-5, during the Twenty-Fifth Anniversary year of the journal.

The Official website of the Fibonacci Association [3] The Fibonacci Quarterly [1] Up-to-date list of issues of The Fibonacci Quarterly [4]

References
[1] [2] [3] [4] http:/ / www. fq. math. ca/ http:/ / www. mscs. dal. ca/ Fibonacci/ fourteenth1. pdf http:/ / www. mscs. dal. ca/ Fibonacci/ index. html http:/ / www. fq. math. ca/ list-of-issues. html

Fibonacci Quarterly

43

Fibonacci Quarterly
As the primary publication of The Fibonacci Association, The Fibonacci Quarterly provides a focus for worldwide interest in the Fibonacci number sequence and related mathematics. Published since 1963, its founding editors were Verner Emil Hoggatt, Jr. and Alfred Brousseau.[1] The present editor is Professor Curtis Cooper of the Mathematics Department of the University of Central Missouri. The Fibonacci Quarterly has an editorial board of nineteen members and is overseen by the nine-member board of directors of The Fibonacci Association. The journal includes research articles, expository articles, Elementary Problems and Solutions, Advanced Problems and Solutions, and announcements of interest to members of The Fibonacci Association. Occasionally, the journal publishes special invited articles by distinguished mathematicians. An online Index to The Fibonacci Quarterly [2] covering Volumes 1-48 (1963-2010) includes a Title Index, Author Index, Elementary Problem Index, Advanced Problem Index, Miscellaneous Problem Index, and Quick Reference Keyword Index. The Fibonacci Quarterly is available online to subscribers; on June 29, 2010, online volumes ranged from the current issue back to volume 41 (2003). Many articles in The Fibonacci Quarterly deal directly with topics that are very closely related to Fibonacci numbers, such as Lucas numbers, the golden ratio, Zeckendorf representations, Binet forms, Fibonacci polynomials, and Chebyshev polynomials. However, many other topics, especially as related to recurrences, are also well represented. These include primes, pseudoprimes, graph colorings, Euler numbers, continued fractions, Stirling numbers, Pythagorean triples, Ramsey theory, Lucas-Bernoulli numbers, quadratic residues, higher-order recurrence sequences, nonlinear recurrence sequences, combinatorial proofs of number-theoretic identities, Diophantine equations, special matrices and determinants, the Collatz sequence, public-key crypto functions, elliptic curves, fractal dimension, hypergeometric functions, Fibonacci polytopes, geometry, graph theory, music, and art.

Notes
[1] Biography of Verner Emil Hoggatt, Jr. (http:/ / faculty. evansville. edu/ ck6/ bstud/ hoggatt. html) by Clark Kimberling [2] http:/ / www. mscs. dal. ca/ Fibonacci/ quarterlyindex. html

References
Bicknell-Johnson, Marjorie (1987), "A short history of The Fibonacci Quarterly", The Fibonacci Quarterly 25: 25.

The Fibonacci Quarterly (http://www.fq.math.ca/) Homepage Up-to-date list of issues of The Fibonacci Quarterly (http://www.fq.math.ca/list-of-issues.html) Fibonacci: Eight Hundred Years Young (http://faculty.evansville.edu/ck6/bstud/fibo.html) by A. F. Horadam Biography of Brother Alfred Brousseau, F. S. C. (http://faculty.evansville.edu/ck6/bstud/brousseau.html) by Clark Kimberling

Fibonacci numbers

44

Fibonacci numbers
In mathematics, the Fibonacci numbers are the numbers in the following integer sequence:

A tiling with squares whose sides are successive Fibonacci numbers in length

A yupana (Quechua for "counting tool") is a calculator that was used by the Incas. Researchers believe that calculations were based on Fibonacci numbers to minimize the number of necessary [1] grains per field.

A Fibonacci spiral created by drawing arcs connecting the opposite corners of squares in the Fibonacci tiling; this one uses squares of sizes 1, 1, 2, 3, 5, 8, 13, 21, and 34; see Golden spiral

(sequence A000045 [2] in OEIS) By definition, the first two Fibonacci numbers are 0 and 1, and each subsequent number is the sum of the previous two. Some sources omit the initial 0, instead beginning the sequence with two 1s. In mathematical terms, the sequence Fn of Fibonacci numbers is defined by the recurrence relation with seed values

Fibonacci numbers

45

The Fibonacci sequence is named after Leonardo of Pisa, who was known as Fibonacci (a contraction of filius Bonacci, "son of Bonaccio"). Fibonacci's 1202 book Liber Abaci introduced the sequence to Western European mathematics, although the sequence was independently described in Indian mathematics and it is disputed which came first. Fibonacci numbers are used in the analysis of financial markets, in strategies such as Fibonacci retracement, and are used in computer algorithms such as the Fibonacci search technique and the Fibonacci heap data structure. The simple recursion of Fibonacci numbers has also inspired a family of recursive graphs called Fibonacci cubes for interconnecting parallel and distributed systems. They also appear in biological settings,[3] such as branching in trees, arrangement of leaves on a stem, the fruit spouts of a pineapple,[4] the flowering of artichoke, an uncurling fern and the arrangement of a pine cone.[5]

Origins
The Fibonacci sequence was known in Indian mathematics independently of the West, but scholars differ on the timing of its discovery. Susantha Goonatilake writes that "Its development is attributed in part to Pingala (200 BC), later being associated with Virahanka (c. 700 AD), Gopla (c.1135 AD), and Hemachandra (c.1150)."[6] Parmanand Singh cites Pingala's contributions to the analysis of prosody, but writes that Virahanka was "the first authority who explicitly gave the rule for the formation" of the Fibonacci numbers.[7] In contrast, Rachel Hall only mentions Hemachandra among these authors as having worked with Fibonacci numbers; she claims that around 1150, Hemachandra noticed that the number of possible rhythms followed the Fibonacci sequence.[8] According to Donald Knuth, the motivation for the Indian study of these numbers came from Sanskrit prosody, where long syllables have duration 2 and short syllables have duration 1. As Knuth observes, Pingala numbered the different sequences that may be formed by these syllables, and Kedara (8th century AD) gave procedures for generating them all, but the connection to the Fibonacci sequence comes more specifically from considering the sequences with a fixed total duration, for which Knuth cites the anonymous Prakrta Paingala (c. 1320 AD). The number of sequences with duration m is the Fibonacci number Fm+1, and the techniques in Prakrta Paingala involve Fibonacci coding.[9]

In the West, the sequence was studied by Leonardo of Pisa, known as Fibonacci, in his Liber Abaci (1202).[10] He considers the growth of an idealized (biologically unrealistic) rabbit population, assuming that: a newly-born pair of rabbits, one male, one female, are put in a field; rabbits are able to mate at the age of one month so that at the end of its second month a female can produce another pair of rabbits; rabbits never die and a mating pair always produces one new pair (one male, one female) every month from the second month on. The puzzle that Fibonacci posed was: how many pairs will there be in one year? At the end of the first month, they mate, but there is still only 1 pair. At the end of the second month the female produces a new pair, so now there are 2 pairs of rabbits in the field. At the end of the third month, the original female produces a second pair, making 3 pairs in all in the field. At the end of the fourth month, the original female has produced yet another new pair, the female born two months ago produces her first pair also, making 5 pairs.

At the end of the nth month, the number of pairs of rabbits is equal to the number of new pairs (which is the number of pairs in month n-2) plus the number of pairs alive last month. This is the nth Fibonacci number.[11]

Fibonacci numbers

46

## List of Fibonacci numbers

The first 50 Fibonacci numbers Fn for n=0,1,2, ... ,49 are:[12] [13]
F0= F5= F10= F15= F20= F25= F30= F35= F40= 0 5 F1= F6= 1 8 F2= F7= 1 13 F3= F8= 2 21 F4= F9= 3 34 377 4,181 46,368 514,229 5,702,887 63,245,986 701,408,733

55 F11= 610 F16= 6,765 F21= 75,025 F26= 832,040 F31= 9,227,465 F36= 102,334,155 F41=

89 F12= 987 F17= 10,946 F22= 121,393 F27= 1,346,269 F32= 14,930,352 F37= 165,580,141 F42=

144 F13= 1,597 F18= 17,711 F23= 196,418 F28= 2,178,309 F33= 24,157,817 F38= 267,914,296 F43=

233 F14= 2,584 F19= 28,657 F24= 317,811 F29= 3,524,578 F34= 39,088,169 F39= 433,494,437 F44=

F45= 1,134,903,170 F46= 1,836,311,903 F47= 2,971,215,073 F48= 4,807,526,976 F49= 7,778,742,049

The sequence can also be extended to negative index n using the re-arranged recurrence relation

## Thus the complete sequence is

Properties
Every 3rd number of the sequence is even and more generally, every kth number of the sequence is a multiple of Fk. Thus the Fibonacci sequence is an example of a divisibility sequence. In fact, the Fibonacci sequence satisfies the stronger divisibility property

The Fibonacci numbers are also an example of a complete sequence. This means that every positive integer can be written as a sum of Fibonacci numbers, where any one number is used once at most.

Fibonacci numbers

47

## Relation to the golden ratio

Closed-form expression
Like every sequence defined by linear recurrence,[14] the Fibonacci numbers have a closed-form solution. It has become very well known as Binet's formula, even though it was already known by Abraham de Moivre:[15]

Approximate and true golden spirals. The green spiral is made from quarter-circles tangent to the interior of each square, while the red spiral is a Golden Spiral, a special type of logarithmic spiral. Overlapping portions appear yellow. The length of the side of one square divided by that of the next smaller square is the golden ratio.

where

## which is also known as the generating polynomial of the recursion.

Fibonacci numbers Computation by rounding Since for all , the number is the closest integer to Therefore it

48

## can be found by rounding, or in terms of the floor function:

Similarly, if we already know that the number F is a Fibonacci number, we can determine its index within the sequence by

## Limit of consecutive quotients

Johannes Kepler observed that the ratio of consecutive Fibonacci numbers converges. He wrote that "as 5 is to 8 so is 8 to 13, practically, and as 8 is to 13, so is 13 to 21 almost, and concluded that the limit approaches the golden ratio .[17]

This convergence does not depend on the starting values chosen, excluding 0, 0. For example, the initial values 19 and 31 generate the sequence 19, 31, 50, 81, 131, 212, 343, 555 ... etc. The ratio of consecutive terms in this sequence shows the same convergence towards the golden ratio. In general, we have:

## Decomposition of powers of the golden ratio

Since the golden ratio satisfies the equation

this expression can be used to decompose higher powers Fibonacci numbers as the linear coefficients: This expression is also true for Fibonacci rule

as a linear function of lower powers, which in turn can and 1. The resulting recurrence relationships yield

## is extended to negative integers using the

Matrix form
A 2-dimensional system of linear difference equations that describes the Fibonacci sequence is

or

and

and

## Using these facts, and the properties of eigenvalues, we can derive a

Fibonacci numbers direct formula for the nth element in the fibonacci series:

49

The matrix has a determinant of 1, and thus it is a 22 unimodular matrix. This property can be understood in terms of the continued fraction representation for the golden ratio:

The Fibonacci numbers occur as the ratio of successive convergents of the continued fraction for formed from successive convergents of any continued fraction has a determinant of +1 or 1. The matrix representation gives the following closed expression for the Fibonacci numbers:

## , and the matrix

Taking the determinant of both sides of this equation yields Cassini's identity

## for any square matrix A, the following identities can be derived:

In particular, with

## Recognizing Fibonacci numbers

The question may arise whether a positive integer z is a Fibonacci number. Since , the most straightforward, brute-force test is the identity is the closest integer to

which is true if and only if z is a Fibonacci number. In this formula, previously discussed closed-form expressions.

## Alternatively, a positive integer z is a Fibonacci number if and only if one of square.

[19]

A slightly more sophisticated test uses the fact that the convergents of the continued fraction representation of ratios of successive Fibonacci numbers. That is the inequality

(with coprime positive integers p, q) is true if and only if p and q are successive Fibonacci numbers. From this one derives the criterion that z is a Fibonacci number if and only if the closed interval

Fibonacci numbers contains a positive integer.[20] For , it is easy to show that this interval contains at most one integer, and in

50

the event that z is a Fibonacci number, the contained integer is equal to the next successive Fibonacci number after z. Somewhat remarkably, this result still holds for the case , but it must be stated carefully since appears twice in the Fibonacci sequence, and thus has two distinct successors.

Identities
Most identities involving Fibonacci numbers draw from combinatorial arguments. F(n) can be interpreted as the number of sequences of 1s and 2s that sum to n 1, with the convention that F(0) = 0, meaning no sum will add up to 1, and that F(1) = 1, meaning the empty sum will "add up" to 0. Here the order of the summands matters. For example, 1 + 2 and 2 + 1 are considered two different sums and are counted twice. This is discussed in further detail at YoungFibonacci lattice.

## The YoungFibonacci graph, showing a combinatorial interpretation of the Fibonacci numbers.

First identity
For n > 1. The nth Fibonacci number is the sum of the previous two Fibonacci numbers.

Second identity

The sum of the first n Fibonacci numbers is the (n+2)nd Fibonacci number minus 1.

Third identity
This identity has slightly different forms for Fj, depending on whether j is odd or even. The sum of the first n1 Fibonacci numbers, Fj, such that j is odd, is the (2n)th Fibonacci number.

The sum of the first n Fibonacci numbers, Fj, such that j is even, is the (2n+1)th Fibonacci number minus 1.

[21]

Fibonacci numbers

51

Fourth identity

Fifth identity

The sum of the squares of the first n Fibonacci numbers is the product of the nth and (n+1)th Fibonacci numbers.

## Identity for doubling n

[15]

Another identity
Another identity useful for calculating Fn for large values of n is[15]

from which other identities for specific values of k, n, and c can be derived below, including for all integers n and k. Dijkstra[18] points out that doubling identities of this type can be used to calculate Fn using O(logn) long multiplication operations of size n bits. The number of bits of precision needed to perform each multiplication doubles at each step, so the performance is limited by the final multiplication; if the fast Schnhage-Strassen multiplication algorithm is used, this is O(nlognloglogn) bit operations. Notice that, with the definition of Fibonacci numbers with negative n given in the introduction, this formula reduces to the double n formula when k=0.

Other identities
Other identities include relationships to the Lucas numbers, which have the same recursive properties but start with L0=2 and L1=1. These properties include F2n=FnLn. There are also scaling identities, which take you from Fn and Fn+1 to a variety of things of the form Fan+b; for instance by Cassini's identity.

These can be found experimentally using lattice reduction, and are useful in setting up the special number field sieve to factorize a Fibonacci number. Such relations exist in a very general sense for numbers defined by recurrence relations. See the section on multiplication formulae under Perrin numbers for details.

Fibonacci numbers

52

Power series
The generating function of the Fibonacci sequence is the power series

## This series has a simple and interesting closed-form solution for

This solution can be proven by using the Fibonacci recurrence to expand each coefficient in the infinite sum defining :

for

## for all integers Conversely,

Reciprocal sums
Infinite sums over reciprocal Fibonacci numbers can sometimes be evaluated in terms of theta functions. For example, we can write the sum of every odd-indexed reciprocal Fibonacci number as

## and the sum of squared reciprocal Fibonacci numbers as

If we add 1 to each Fibonacci number in the first sum, there is also the closed form

and there is a nice nested sum of squared Fibonacci numbers giving the reciprocal of the golden ratio,

Fibonacci numbers

53

Results such as these make it plausible that a closed formula for the plain sum of reciprocal Fibonacci numbers could be found, but none is yet known. Despite that, the reciprocal Fibonacci constant

has been proved irrational by Richard Andr-Jeannin. Millin series gives a remarkable identity:[23]

which follows from the closed form for its partial sums as N tends to infinity:

## Primes and divisibility

Fibonacci primes
A Fibonacci prime is a Fibonacci number that is prime (sequence A005478 [1] in OEIS). The first few are: 2, 3, 5, 13, 89, 233, 1597, 28657, 514229, ... Fibonacci primes with thousands of digits have been found, but it is not known whether there are infinitely many.[24] Fkn is divisible by Fn, so, apart from F4 = 3, any Fibonacci prime must have a prime index. As there are arbitrarily long runs of composite numbers, there are therefore also arbitrarily long runs of composite Fibonacci numbers. With the exceptions of 1, 8 and 144 (F1 = F2, F6 and F12) every Fibonacci number has a prime factor that is not a factor of any smaller Fibonacci number (Carmichael's theorem).[25] 144 is the only nontrivial square Fibonacci number.[26] Attila Peth proved[27] in 2001 that there are only finitely many perfect power Fibonacci numbers. In 2006, Y. Bugeaud, M. Mignotte, and S. Siksek proved that only 8 and 144 are non-trivial perfect powers.[28] No Fibonacci number greater than F6 = 8 is one greater or one less than a prime number.[29] Any three consecutive Fibonacci numbers, taken two at a time, are relatively prime: that is, gcd(Fn, Fn+1) = gcd(Fn, Fn+2) = 1. More generally, gcd(Fn, Fm) = Fgcd(n, m).[30] [31]

## Prime divisors of Fibonacci numbers

The divisibility of Fibonacci numbers by a prime p is related to the Legendre symbol follows: which is evaluated as

## If p is a prime number then For example,

[32] [33]

Fibonacci numbers

54

It is not known whether there exists a prime p such that would be called Wall-Sun-Sun primes. Also, if p 5 is an odd prime number then:[34]

## Examples of all the cases:

For odd n, all odd prime divisors of Fn are 1(mod4), implying that all odd divisors of Fn (as the products of odd prime divisors) are 1(mod4).[35] [36] For example, F1 = 1, F3 = 2, F5 = 5, F7 = 13, F9 = 34 = 217, F11 = 89, F13 = 233, F15 = 610 = 2561

Fibonacci numbers

55

Divisibility by 11
The sum of any 10 consecutive Fibonacci numbers is divisible by 11; i.e.,

## For example, let n = 1:

n = 2:

n = 3:

In fact, the identity is true for all integers n, not just positive ones: n = 0:

n = 1:

n = 2:

Periodicity modulo n
It may be seen that if the members of the Fibonacci sequence are taken modn, the resulting sequence must be periodic with period at mostn2. The lengths of the periods for various n form the so-called Pisano periods (sequence A001175 [1] in OEIS). Determining the Pisano periods in general is an open problem, although for any particular n it can be solved as an instance of cycle detection.

Right triangles
Starting with 5, every second Fibonacci number is the length of the hypotenuse of a right triangle with integer sides, or in other words, the largest number in a Pythagorean triple. The length of the longer leg of this triangle is equal to the sum of the three sides of the preceding triangle in this series of triangles, and the shorter leg is equal to the difference between the preceding bypassed Fibonacci number and the shorter leg of the preceding triangle. The first triangle in this series has sides of length 5, 4, and 3. Skipping 8, the next triangle has sides of length 13, 12 (5+4+3), and 5 (83). Skipping 21, the next triangle has sides of length 34, 30 (13+12+5), and 16 (215). This series continues indefinitely. The triangle sides a, b, c can be calculated directly:

## for all n, but they only represent triangle sides when

Any four consecutive Fibonacci numbers Fn, Fn+1, Fn+2 and Fn+3 can also be used to generate a Pythagorean triple in a different way:

## Example 1: let the Fibonacci numbers be 1, 2, 3 and 5. Then:

Fibonacci numbers

56

## Example 2: let the Fibonacci numbers be 8, 13, 21 and 34. Then:

Magnitude
Since is asymptotic to , the number of digits in is asymptotic to . As a consequence, for every integer there are either 4 or 5 Fibonacci numbers with d decimal digits. More generally, in the base b representation, the number of digits in is asymptotic to .

## Binary strings and compositions

The Fibonacci numbers can be found in different ways in the sequence of binary strings. The number of binary strings of length n without consecutive 1s is the Fibonacci number Fn+2. For example, out of the 16 binary strings of length 4, there are F6 = 8 without consecutive 1s - they are 0000, 1000, 0100, 0010, 1010, 0001, 1001 and 0101. By symmetry, the number of strings of length n without consecutive 0s is also Fn+2. The number of binary strings of length n without an odd number of consecutive 1s is the Fibonacci number Fn+1. For example, out of the 16 binary strings of length 4, there are F5 = 5 without an odd number of consecutive 1s they are 0000, 0011, 0110, 1100, 1111. The number of binary strings of length n without an even number of consecutive 0s or 1s is 2Fn. For example, out of the 16 binary strings of length 4, there are 2F4 = 6 without an even number of consecutive 0s or 1s - they are 0001, 1000, 1110, 0111, 0101, 1010.

Applications
The Fibonacci numbers are important in the computational run-time analysis of Euclid's algorithm to determine the greatest common divisor of two integers: the worst case input for this algorithm is a pair of consecutive Fibonacci numbers.[37] Yuri Matiyasevich was able to show that the Fibonacci numbers can be defined by a Diophantine equation, which led to his original solution of Hilbert's tenth problem. The Fibonacci numbers occur in the sums of "shallow" diagonals in Pascal's triangle and Lozani's triangle (see Binomial coefficient). They occur more obviously in Hosoya's triangle. Every positive integer can be written in a unique way as the sum of one or more distinct Fibonacci numbers in such a way that the sum does not include any two consecutive Fibonacci numbers. This is known as Zeckendorf's theorem, and a sum of Fibonacci numbers that satisfies these conditions is called a Zeckendorf representation. The Zeckendorf representation of a number can be used to derive its Fibonacci coding. The Fibonacci numbers and principle is also used in the financial markets. It is used in trading algorithms, applications and strategies. Some typical forms include: the Fibonacci fan, the Fibonacci arc, Fibonacci retracement and the Fibonacci time extension.[38] Fibonacci numbers are used by some pseudorandom number generators.

Fibonacci numbers Fibonacci numbers are used in a polyphase version of the merge sort algorithm in which an unsorted list is divided into two lists whose lengths correspond to sequential Fibonacci numbers - by dividing the list so that the two parts have lengths in the approximate proportion . A tape-drive implementation of the polyphase merge sort was described in The Art of Computer Programming. Fibonacci numbers arise in the analysis of the Fibonacci heap data structure. The Fibonacci cube is an undirected graph with a Fibonacci number of nodes that has been proposed as a network topology for parallel computing. A one-dimensional optimization method, called the Fibonacci search technique, uses Fibonacci numbers.[39] The Fibonacci number series is used for optional lossy compression in the IFF 8SVX audio file format used on Amiga computers. The number series compands the original audio wave similar to logarithmic methods e.g. -law.[40] [41] In music, Fibonacci numbers are sometimes used to determine tunings, and, as in visual art, to determine the length or size of content or formal elements. It is commonly thought that the third movement of Bla Bartk's Music for Strings, Percussion, and Celesta was structured using Fibonacci numbers. Since the conversion factor 1.609344 for miles to kilometers is close to the golden ratio (denoted ), the decomposition of distance in miles into a sum of Fibonacci numbers becomes nearly the kilometer sum when the Fibonacci numbers are replaced by their successors. This method amounts to a radix 2 number register in golden ratio base being shifted. To convert from kilometers to miles, shift the register down the Fibonacci sequence instead.[42]

57

In nature
Fibonacci sequences appear in biological settings,[3] in two consecutive Fibonacci numbers, such as branching in trees, arrangement of leaves on a stem, the fruitlets of a pineapple,[4] the flowering of artichoke, an uncurling fern and the arrangement of a pine cone.[5] In addition, numerous poorly substantiated claims of Fibonacci numbers or golden sections in nature are found in popular sources, e.g. relating to the breeding of rabbits, the spirals of shells, and the curve of waves.[43] The Fibonacci numbers are also found in the family tree of honeybees.[44]
Sunflower head displaying florets in spirals of 34 and 55 around the outside

Fibonacci numbers

58

Przemysaw Prusinkiewicz advanced the idea that real instances can in part be understood as the expression of certain algebraic constraints on free groups, specifically as certain Lindenmayer grammars.[45] A model for the pattern of florets in the head of a sunflower was proposed by H. Vogel in 1979.[46] This has the form

## The shell of Achatina fulica respects the fibonacci logic

where n is the index number of the floret and c is a constant scaling factor; the florets thus lie on Fermat's spiral. The divergence angle, approximately 137.51, is the golden angle, dividing the circle in the golden ratio. Because this ratio is irrational, no floret has a neighbor at exactly the same angle from the center, so the florets pack efficiently. Because the rational approximations to the golden ratio are of the form F(j):F(j+1), the nearest neighbors of floret number n are those at nF(j) for some index j which depends on r, the distance from the center. It is often said that sunflowers and similar arrangements have 55 spirals in one direction and 89 in the other (or some other pair of adjacent Fibonacci numbers), but this is true only of one range of radii, typically the outermost and thus most conspicuous.[47]

## The bee ancestry code

Fibonacci numbers also appear in the description of the reproduction of a population of idealized honeybees, according to the following rules: If an egg is laid by an unmated female, it hatches a male or drone bee. If, however, an egg was fertilized by a male, it hatches a female. Thus, a male bee will always have one parent, and a female bee will have two. If one traces the ancestry of any male bee (1 bee), he has 1 female parent (1 bee). This female had 2 parents, a male and a female (2 bees). The female had two parents, a male and a female, and the male had one female (3 bees). Those two females each had two parents, and the male had one (5 bees). This sequence of numbers of parents is the Fibonacci sequence.[48] This is an idealization that does not describe actual bee ancestries. In reality, some ancestors of a particular bee will always be sisters or brothers, thus breaking the lineage of distinct parents.

Fibonacci numbers

59

Generalizations
The Fibonacci sequence has been generalized in many ways. These include: Generalizing the index to negative integers to produce the Negafibonacci numbers. Generalizing the index to real numbers using a modification of Binet's formula.[15] Starting with other integers. Lucas numbers have L1 = 1, L2 = 3, and Ln = Ln1 + Ln2. Primefree sequences use the Fibonacci recursion with other starting points in order to generate sequences in which all numbers are composite. Letting a number be a linear function (other than the sum) of the 2 preceding numbers. The Pell numbers have Pn = 2Pn 1 + Pn 2. Not adding the immediately preceding numbers. The Padovan sequence and Perrin numbers have P(n) = P(n 2) + P(n 3). Generating the next number by adding 3 numbers (tribonacci numbers), 4 numbers (tetranacci numbers), or more. Adding other objects than integers, for example functions or stringsone essential example is Fibonacci polynomials.

Notes
[1] http:/ / www. quipus. it/ english/ Andean%20Calculators. pdf [2] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa000045 [3] S. Douady and Y. Couder (1996). "Phyllotaxis as a Dynamical Self Organizing Process" (http:/ / www. math. ntnu. no/ ~jarlet/ Douady96. pdf) (PDF). Journal of Theoretical Biology 178 (178): 255274. doi:10.1006/jtbi.1996.0026. . [4] Jones, Judy; William Wilson (2006). "Science". An Incomplete Education. Ballantine Books. p.544. ISBN978-0-7394-7582-9. [5] A. Brousseau (1969). "Fibonacci Statistics in Conifers". Fibonacci Quarterly (7): 525532. [6] Susantha Goonatilake (1998). Toward a Global Science (http:/ / books. google. com/ ?id=SI5ip95BbgEC& pg=PA126& dq=Virahanka+ Fibonacci). Indiana University Press. p.126. ISBN9780253333889. . [7] Singh, Parmanand (1985). "The So-called Fibonacci numbers in ancient and medieval India". Historia Mathematica 12 (3): 229244. doi:10.1016/0315-0860(85)90021-7. [8] Hall, Rachel W. (2008). "Math for poets and drummers" (http:/ / www. sju. edu/ ~rhall/ mathforpoets. pdf). Math Horizons 15: 1011. . [9] Donald Knuth (2006). The Art of Computer Programming: Generating All TreesHistory of Combinatorial Generation; Volume 4 (http:/ / books. google. com/ ?id=56LNfE2QGtYC& pg=PA50& dq=rhythms). Addison-Wesley. p.50. ISBN9780321335708. . [10] Sigler, Laurence E. (trans.) (2002). Fibonacci's Liber Abaci. Springer-Verlag. ISBN0-387-95419-8. Chapter II.12, pp. 404405. [11] Knott, Ron. "Fibonacci's Rabbits" (http:/ / www. maths. surrey. ac. uk/ hosted-sites/ R. Knott/ Fibonacci/ fibnat. html#Rabbits). University of Surrey Faculty of Engineering and Physical Sciences. . [12] By modern convention, the sequence begins with F0=0. The Liber Abaci began the sequence with F1 = 1, omitting the initial 0, and the sequence is still written this way by some. [13] The website (http:/ / www. maths. surrey. ac. uk/ hosted-sites/ R. Knott/ Fibonacci/ fibtable. html) has the first 300 Fn factored into primes and links to more extensive tables. [14] Greene, Daniel H.; Knuth, Donald E. (1982). "2.1.1 Constant coefficients A) Homogeneous equations". Mathematics for the Analysis of Algorithms (2nd ed.). Birkhuser. p. 17. [15] Weisstein, Eric W., " Fibonacci Number (http:/ / mathworld. wolfram. com/ FibonacciNumber. html)" from MathWorld. [16] http:/ / en. wikipedia. org/ wiki/ Oeis%3Aa001622 [17] Kepler, Johannes (1966). A New Year Gift: On Hexagonal Snow. Oxford University Press. p.92. ISBN0198581203. Strena seu de Nive Sexangula (1611). [18] E. W. Dijkstra (1978). In honour of Fibonacci. Report EWD654 (http:/ / www. cs. utexas. edu/ users/ EWD/ ewd06xx/ EWD654. PDF) [19] Posamentier, Alfred; Lehmann, Ingmar (2007). The (Fabulous) FIBONACCI Numbers. Prometheus Books. p.305. ISBN978-1-59102-475-0. [20] M.Mbius, Wie erkennt man eine Fibonacci Zahl?, Math. Semesterber. (1998) 45; 243246. [21] Vorobiev, Nikola Nikolaevich; Mircea Martin (2002). "Chapter 1". Fibonacci Numbers. Birkhuser. pp.56. ISBN3-7643-6135-2. [22] The Remarkable Number 1/89 (http:/ / www. geom. uiuc. edu/ ~rminer/ 1over89/ ) at The Geometry Center. [23] Weisstein, Eric W., " Millin Series (http:/ / mathworld. wolfram. com/ MillinSeries. html)" from MathWorld. [24] Weisstein, Eric W., " Fibonacci Prime (http:/ / mathworld. wolfram. com/ FibonacciPrime. html)" from MathWorld. [25] Ron Knott, "The Fibonacci numbers" (http:/ / www. maths. surrey. ac. uk/ hosted-sites/ R. Knott/ Fibonacci/ fibtable. html). [26] J H E Cohn (1964). "Square Fibonacci Numbers Etc" (http:/ / math. la. asu. edu/ ~checkman/ SquareFibonacci. html). Fibonacci Quarterly 2: 109113. . [27] A. Peth, Diophantine properties of linear recursive sequences II, Acta Math. Paedagogicae Nyregyhziensis, 17(2001), 8196.

Fibonacci numbers
[28] Y. Bugeaud, M. Mignotte, S. Siksek: Classical and modular approaches to exponential Diophantine equations. I. Fibonacci and Lucas perfect powers. Ann. of Math. (2), 163(2006), 9691018. [29] Ross Honsberger Mathematical Gems III (AMS Dolciani Mathematical Expositions No. 9), 1985, ISBN 0-88385-318-3, p. 133. [30] Paulo Ribenboim, My Numbers, My Friends, Springer-Verlag 2000. [31] Su, Francis E., et al. "Fibonacci GCD's, please." (http:/ / www. math. hmc. edu/ funfacts/ ffiles/ 20004. 5. shtml), Mudd Math Fun Facts. [32] Paulo Ribenboim (1996), The New Book of Prime Number Records, New York: Springer, ISBN 0-387-94457-5, p. 64. [33] Franz Lemmermeyer (2000), Reciprocity Laws, New York: Springer, ISBN 3-540-66957-4, ex 2.252.28, pp. 7374. [34] Lemmermeyer, ex. 2.28, pp. 7374. [35] Lemmermeyer, ex. 2.27 p. 73. [36] The website (http:/ / www. maths. surrey. ac. uk/ hosted-sites/ R. Knott/ Fibonacci/ fibtable. html) has the first 300 Fibonacci numbers factored into primes. [37] Knuth, Donald E. (1997). The Art of Computer Programming, Volume 1: Fundamental Algorithms (3rd ed.). Addison-Wesley. ISBN0-201-89683-4. (p. 343). [38] http:/ / www. investopedia. com/ terms/ f/ fibonaccilines. asp [39] M. Avriel and D.J. Wilde (1966). "Optimality of the Symmetric Fibonacci Search Technique". Fibonacci Quarterly (3): 265269. [40] Amiga ROM Kernel Reference Manual, Addison-Wesley 1991. [41] IFF - MultimediaWiki (http:/ / wiki. multimedia. cx/ index. php?title=IFF#Fibonacci_Delta_Compression). [42] Zeckendorf representation (http:/ / eom. springer. de/ Z/ z120020. htm). [43] "Fibonacci Flim-Flam" (http:/ / www. lhup. edu/ ~dsimanek/ pseudo/ fibonacc. htm). . [44] "Marks for the da Vinci Code: B" (http:/ / www. cs4fn. org/ maths/ bee-davinci. php). Computer Science For Fun: CS4FN. . [45] Prusinkiewicz, Przemyslaw; James Hanan (1989). Lindenmayer Systems, Fractals, and Plants (Lecture Notes in Biomathematics). Springer-Verlag. ISBN0-387-97092-4. [46] Vogel, H (1979). "A better way to construct the sunflower head". Mathematical Biosciences 44 (44): 179189. doi:10.1016/0025-5564(79)90080-4 [47] Prusinkiewicz, Przemyslaw; Lindenmayer, Aristid (1990). [[The Algorithmic Beauty of Plants (http:/ / algorithmicbotany. org/ papers/ #webdocs)]]. Springer-Verlag. pp.101107. ISBN978-0387972978. . [48] The Fibonacci Numbers and the Ancestry of Bees (http:/ / www1. math. american. edu/ newstudents/ shared/ puzzles/ fibbee. html).

60

Sequence A000045 (http://en.wikipedia.org/wiki/Oeis:a000045) Fibonacci Numbers at the On-Line Encyclopedia of Integer Sequences Periods of Fibonacci Sequences Mod m (http://www.mathpages.com/home/kmath078.htm) at MathPages Chaotic behaviour in Fibonacci sequences (http://www.scientificblogging.com/hammock_physicist/ fibonacci_butterflies) Scientists find clues to the formation of Fibonacci spirals in nature (http://www.physorg.com/news97227410. html) Find the n-th Fibonacci number using a generator (http://81.167.229.63/~srbzorg/fibonaccigenerator.php) Implementations to calculate the Fibonacci sequence in 21 languages (http://codecodex.com/wiki/ Calculate_the_Fibonacci_sequence)

61

62

63