This action might not be possible to undo. Are you sure you want to continue?

l

Lecture 3 Gaussian Probability Distribution

l

Gaussian probability distribution is perhaps the most used distribution in all of science. u also called “bell shaped curve” or normal distribution Unlike the binomial and Poisson distribution, the Gaussian is a continuous distribution:

P(y) = 1 e s 2p

( y-m) 2 2s 2

†

l

m = mean of distribution (also at the same place as mode and median) s2 = variance of distribution y is a continuous variable (-∞ £ y £ ∞) Probability (P) of y being in the range [a, b] is given by an integral:

P(a < y < b) =

u

1 Úe s 2p a

b -

( y-m) 2 2s 2 dy

Karl Friedrich Gauss 1777-1855 The integral for arbitrary a and b cannot be evaluated analytically + The value of the integral has to be looked up in a table (e.g. Appendixes A and B of Taylor).

†

P(x) Plot of Gaussian pdf x K.K. Gan L3: Gaussian Probability Distribution 1

1 p(x) = e s 2p (x - m )2 2 2s

gaussian

+ We can associate a probability for a measurement to be |m .l The total area under the curve is normalized to one. of exceeding ±ns 0. Gan L3: Gaussian Probability Distribution 2 .32 measurement taken at random from a 2 0.003 from the true mean of the distribution.3%) that a 1 0.00006 P(-• < y < •) = • ( y-m) 2 2s 2 dy = 1 l † Relationship between Gaussian and Binomial distribution l l The Gaussian distribution can be derived from the binomial (or Poisson) assuming: u p is finite u N is very large u we have a continuous variable rather than a discrete variable An example illustrating the small difference between the two distributions under the above conditions: u Consider tossing a coin 10. ns Prob.67 0. + the probability integral: 1 Úe s 2p -• We often talk about a measurement being a certain number of standard deviations (s) away from the mean (m) of the Gaussian. 4 0.ns| from the mean just by calculating the area outside of this region.K.5 It is very unlikely (< 0. p(heads) = 0.000 time.000 K.5 N = 10.05 Gaussian pdf will be more than ± 3s 3 0.

.s < y < m + s ) = Ú e 2s dy ª 0. A more exact statement: Actually.K.Yn be an infinite sequence of independent random variables be from different pdf’s! each with the same probability distribution..69 Â 4 m=5000-50 (10 .. Gan L3: Gaussian Probability Distribution 3 .p)]1/2 = 50 + The probability to be within ±1s for this binomial distribution is: 5000+50 4 10 4 ! P= 0.L.510 -m = 0. u Suppose that the mean (m) and variance (s2) of this distribution are both finite.T. Y2. † K.m)!m! n For a Gaussian distribution: ( y-m) 2 m+s 2 1 P(m . the Y’s can u Let Y1. tells us that under a wide range of circumstances the probability distribution that describes the sum of random variables tends towards a Gaussian distribution as the number of terms in the sum Æ∞.For a binomial distribution: mean number of heads = m = Np = 5000 standard deviation s = [Np(1 .nm ˘ 1 b -1 y2 lim PÍa < 1 2 < b˙ = Ú e 2 dy ˚ nÆ• Î s n 2p a + C.Yn .. + For any numbers a and b: È Y +Y + ..68 s 2p m-s † Both distributions give about the same probability! + n Central Limit Theorem l l l Gaussian distribution is important because of the Central Limit Theorem † A crude statement of the Central Limit Theorem: u Things that are the result of the addition of lots of small effects tend to become Gaussian.5 m 0.

l Recall if y is described by a Gaussian pdf with m = 0 and s = 1 then the probability that a < y < b is given by: + 1 b -1 y2 P ( a < y < b) = Ú e 2 dy 2p a l The CLT is true even if the Y’s are from different pdf’s as long as the means and variances are defined for each pdf! u See Appendix of Barlow for a proof of the Central Limit Theorem. u Devore: Probability and Statistics for Engineering and the Sciences: + A random variable is any rule that associates a number with each outcome in S n S is the set of possible outcomes. Gan L3: Gaussian Probability Distribution 4 .K.Alternatively: È Y -m ˘ È ˘ Y -m 1 b -1 y2 lim PÍa < < b˙ = lim PÍa < < b˙ = Ú e 2 dy nÆ• Î ˚ nÆ• Î s/ n sm 2p a ˚ n sm is sometimes called “the error in the mean” (more on that later). K. l For CLT to be valid: u m and s of pdf must be finite. u l A random variable is not the same as a random number. † No one term in sum should dominate the sum.

l Example: A watch makes an error of at most ±1/2 minute per day. n Since the daily errors come from a uniform distribution with a well defined mean and variance + Central Limit Theorem is applicable: È Y1 +Y2 + .nm 25 . After one year. the average error is zero and the standard deviation 1/√12 minutes.. n The error over the course of a year is just the addition of the daily error.5 .5 1 365 s n 12 † + The probability to be within ± 25 minutes: 1 4.. n For each day.5 † + less than three in a million chance that the watch will be off by more than 25 minutes in a year! † K.Yn .Yn .365 ¥ 0 b= 1 2 = = 4. what’s the probability that the watch is accurate to within ±25 minutes? u Assume that the daily errors are uniform in [-1/2..3¥10 2p -4. 1/2].K.nm ˘ 1 b -1 y2 lim PÍa < < b˙ = Ú e 2 dy ˚ nÆ• Î s n 2p a + The upper limit corresponds to +25 minutes: Y +Y + .Yn .nm -25 ..365 ¥ 0 a= 1 2 = = -4. Gan L3: Gaussian Probability Distribution 5 .1 y 2 -6 P= Ú e 2 dy = 0..999997 = 1.5 1 365 s n 12 † + The lower limit corresponds to -25 minutes: Y +Y + ..

l Example: Generate a Gaussian distribution using random numbers.1] n m = 1/2 and s2 = 1/12 u Procedure: n Take 12 numbers (ri) from your computer’s random number generator n Add them together n Subtract 6 + Get a number that looks as if it is from a Gaussian pdf! È Y +Y2 + . u Random number generator gives numbers distributed uniformly in the interval [0..nm ˘ PÍa < < b˙ Î ˚ s n A) 5000 random numbers B) 5000 pairs (r1 + r2) of random numbers 12 È ˘ 1 Â ri -12 ⋅ 2 Í ˙ i=1 = PÍa < < b˙ 1 ⋅ 12 Í ˙ 12 Í ˙ Î ˚ 12 È ˘ = PÍ-6 < Â ri . Gan L3: Gaussian Probability Distribution 6 . † K.K.6) of random numbers. E E) 5000 12-plets (r1 + r2 +…r12 .Yn .6 < 6˙ Î ˚ C) 5000 triplets (r1 + r2 + r3) D) 5000 12-plets (r1 + r2 +…r12) i=1 1 6 -1 y2 = Ú e 2 dy 2p -6 of random numbers of random numbers. Gaussian m = 0 and s = 1 -6 0 +6 Thus the sum of 12 uniform random numbers minus 6 is distributed as if it came from a Gaussian pdf with m = 0 and s = 1..

Yn .l Example: The daily income of a "card shark" has a uniform distribution in the interval [-$40.4 s n 675 60 201 1 13.16 2p 1 2p 1 16% chance to win > $500 in 60 days L3: Gaussian Probability Distribution 7 † . + need to be normalized in computing the average daily winning (m) and its standard deviation (s).60 ¥ 5 200 a= 1 2 = = =1 s n 675 60 201 The upper limit is the maximum that the shark could win (50$/day for 60 days): Y +Y + . What is the probability that s/he wins more than $500 in 60 days? u Lets use the CLT to estimate this probability: È Y +Y + .nm 3000 ..nm ˘ 1 b -1 y2 lim PÍa < 1 2 < b˙ = Ú e 2 dy ˚ nÆ• Î s n 2p a u The probability distribution of daily income is uniform...60 ¥ 5 2700 b= 1 2 = = = 13.Yn .nm 500 .K.(-40) † u † + K.(-40) 3 ] 3 . Gan The lower limit of the winning is $500: Y +Y + .25 = 675 50 .(-40) =5 2 Ú y p(y)dy s = -40 -m = 50 Ú p(y)dy -40 u 2 1 [50 3 .4 .Yn .. p(y) = 1.(-40)2 ] 50 .1 y 2 1 • -1 y2 P= e 2 dy ª Ú Ú e 2 dy = 0..$50]. 50 † Ú yp(y)dy m = -40 = 50 Ú p(y)dy -40 50 2 1 [50 2 2 ..

- ijwin02
- Chess
- Probability
- Important Tables
- BER simulation using MATLAB
- Bhagavat Geeta
- Borel Sigma
- random variable
- supremum and infimum
- 7.2 Transient
- 9 Laplace z Table
- Laplace Tables
- 8.7 Bessel
- 4.4 Bode
- 6.3 Transfer Func
- 9.5 FIR Filters
- deconvLevinEtalCVPR09
- 22_ofdm2
- Capacity and v-blast Techniques for Mimo Wireless Channel
- Musical Touch Bell
- Basic Concepts in Matlab
- Cognitive Radio Brain Empowered
- Cognitive Radio Brain Empowered
- Digital Signal Processing
- bode_plot

Sign up to vote on this title

UsefulNot useful- Seminar Statistica
- and Standard Normal Distributions (Rev)
- Prob-7
- Normal Distribution
- 7. Continuous Distributions
- Dist
- Chapter 08
- Normal distribution
- Chapter 6
- Normal Probability Distribution
- P416_Lec3_S06
- L6 Normal Distribution
- Exercises Test 6
- Continuous Probability Distribution.pdf
- Normal
- 11BS201 Prob and Stat - Ch33
- stat12t_0602
- Important Probability Distributions
- Handout Five
- Probability
- U1.4-RVDistributions.ppt
- Chap006 New
- The Binomial, Poisson, And Normal Distributions
- Continuous Distributions
- Normal Dist
- Chapter 6
- Chapter 07rger
- 4. Mgf, Statistical Distributions -530
- ASWchapter6
- 4.3
- gaussian

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd