0 Up votes0 Down votes

4 views3 pagesMay 08, 2014

© © All Rights Reserved

PDF, TXT or read online from Scribd

© All Rights Reserved

4 views

© All Rights Reserved

- Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race
- Hidden Figures Young Readers' Edition
- The Law of Explosive Growth: Lesson 20 from The 21 Irrefutable Laws of Leadership
- The Art of Thinking Clearly
- The E-Myth Revisited: Why Most Small Businesses Don't Work and
- The Wright Brothers
- The Other Einstein: A Novel
- State of Fear
- State of Fear
- The Power of Discipline: 7 Ways it Can Change Your Life
- The Kiss Quotient: A Novel
- The 10X Rule: The Only Difference Between Success and Failure
- Being Wrong: Adventures in the Margin of Error
- Algorithms to Live By: The Computer Science of Human Decisions

You are on page 1of 3

1 Some events based on rolls of three dice

(a) = {x

1

x

2

x

2

: 1 x

i

6 for 1 i 3}, or = {111, 112, 113, 114, 115, 116, 121, 122, . . . , 664, 665, 666},

or = {1, 2, 3, 4, 5, 6}

3

.

(b) A = {115, 151, 511, 124, 142, 214, 241, 412, 421, 133, 313, 311, 223, 232, 322}; P(A) =

|A|

||

=

15

6

3

=

5

72

0.0694.

(c) |B| = 6 5 4 because to specify an outcome in B, there are six choices for the rst die, then ve for the

second, and then four for the third. So P(B) =

654

6

3

=

20

36

=

5

9

.

(d) |AB| = |{124, 142, 214, 241, 412, 421}| = 6, so P(AB) =

6

6

3

=

1

36

.

(e) P(A B) = P(A) +P(B) P(AB) =

5

72

+

5

9

1

36

=

5+402

72

=

43

72

.

(f) X takes values in {1, 2, 3}. Now, p

X

(1) = P({111, 222, 333, 444, 555, 666}) =

6

6

3

=

1

36

. (Another way to

see this is that, no matter what shows on the rst die, chances are

1

6

2

that the other two dice will both show

the same number as the rst one.) We found p

X

(3) =

20

36

in part (c) above. So p

X

(2) = 1

1

36

20

36

=

15

36

=

5

12

.

15/36

u

p (u)

0 1 2 3

1/36

X 20/36

2 Concert tickets

(a)

(

5

2

)

(

7

4

)

=

(

5

2

)

(

7

3

)

=

5432

2765

=

2

7

(b)

223

(

7

4

)

=

12

(

7

3

)

=

1232

765

=

12

35

(c)

1

(

5

2

)

=

1

10

3 Using a cumulative distribution function of a mixed type random variable

(a) P{X 10} = F

X

(10) = 1.

(b) P{X 7} = 1 P{X < 7} = 1 F

X

((7)) = 0.75.

(c) P{|X| < 10} = P{X < 10} P{X 10} = F

X

(10) F

X

(10) = 0.75 0.25 = 0.5.

(d) P{X

2

16} = P{4 X 4} = F

X

(4) F

X

(4) = 0.7 0.3 = 0.4.

(e) E[X] =

1

4

(12.5)+

1

2

(0)+

1

4

(10) =

5

8

. (For this solution, we note that the distribution of X is a weighted

mixture of three distributions: uniform distribution over [-15,-10] with weight

1

4

, uniform distribution over

[-5,5] with weight

1

2

, and constant equal to 10 with weight

1

4

.)

4 Scaler quantization of a Gaussian random variable

(a) p

Y

(y) =

1 Q(

2

5

) = Q(

2

5

) y = 1

Q(

1

5

) Q(

2

5

) y = 0.5

1 2Q(

1

5

) y = 0

0 else

(b) Var(Y ) = E[Y

2

] = P[Y {1, 1}] +

1

4

P[Y {0.5, 0.5}] = 2Q(

2

5

) +

1

4

2(Q(

1

5

) Q(

2

5

)) = 1.5Q(

2

5

) +

0.5Q(

1

5

).

5 On the exponential distribution

(a) P{T t} = F

c

T

(t) = e

t

= e

(ln 2)t

= 2

t

. (b) P(T 1|T 2) =

P{T1,T2}

P{T2}

=

P{T1}

P{T2}

=

1

1

2

1

1

4

=

2

3

.

1

6 Standardized version of a random variable

The area under the pdf is one half the base of the triangle times the height, equal to c. So c = 1.

(b) To get mean zero, let a = E[X] = 1. Subtracting a causes the pdf to be centered at zero. And

b =

X

=

1

6

. So

X =

6(X 1). The support of

X is thus the interval (

6,

pdf is the same triangular shape that f

X

has. So the pdf of

X is the following:

1/6

!

6 0 ! #

$

%

& '%(

.

7 Quadratic function of a random variable

(a) Since X ranges over the interval [0, 3], Y ranges over the interval [0, 4].

(b) The expression for F

Y

(c) is qualitatively dierent for 0 c 1 and 1 c 4, as seen in the following

sketch:

c

4

1

1 3

4

1

1 3

c

In each case, F

Y

(c) is equal to one third the length of the shaded interval. For 0 c 1,

F

Y

(c) = P{(X 1)

2

c} = P{1

c X 1 +

c} =

2

c

3

.

For 1 c 4,

F

Y

(c) = P{(X 1)

2

c} = P{0 X 1 +

c} =

1 +

c

3

.

Combining these observations yields:

F

Y

(c) =

0 c < 0

2

c

3

0 c < 1

1+

c

3

1 c < 4

1 c 4

(c) By LOTUS, E[Y ] = E[(X 1)

2

] =

3

0

(u 1)

2 1

3

du = 1

8 A joint distribution

(a) No. For example, the support is not a product set because (0.2, 0.8) and (0.8, 0.2) are in the support of

f

X,Y

but (0.8, 0.8) is not. (b) P{X Y } = P{Y X} by symmetry and P{X Y } + P{Y X} = 1

(because P{X = Y } = 0) so P{X Y } = 0.5. (c) The set {X 0.5, Y 0.5} intersected with the support

of f

X,Y

is the square region shown:

2

v

1

1

0.5

0.5

u

Thus, P{X 0.5, Y 0.5} =

0.5

0

0.5

0

8uv du dv =

1

8

.

(d) By LOTUS, E[

1

XY

] =

1

uv

f

X,Y

(u, v)dudv =

support

8 du dv = 2.

(e) The intersection of {(u, v) : u + v 1} and the support of f

X,Y

is the triangular region {(u, v) : u

0, v 0, u +v 1} as seen in the gure:

1

1

u

v

Integrating the joint density over that region yields:

P{X +Y 1} =

1

0

1u

0

8uv dv du =

1

0

4u(1 u)

2

du

= 4

1

0

(u 2u

2

+u

3

)du = 4

1

2

2

3

+

1

4

=

1

3

9 Some moments

(a) E[X + 2Y + 3Z + 4] = E[X] + 2E[Y ] + 3E[Z] + 4 = 28.

(b) Var(X + 2Y + 3Z + 4) = Var(X) + 2

2

Var(Y ) + 3

2

Var(Z) = (14)(9) = 126.

(c) Cov(X + 2Y, 3Z + 4) = 0.

(d) E[X +Y Z] = E[X] +E[Y ]E[Z] = 20.

(e) E[X

2

] = E[X]

2

+ Var(X) = 25; E[(XY )

2

] = E[X

2

]E[Y

2

] = 625.

(f) E[XY ] = E[X]E[Y ] = 16; Var(XY ) = E[(XY )

2

] E[XY ]

2

= 625 256 = 369.

10 A function of two random variables

(a) P{Z 0.5} = P{Y (0.5)X

2

} =

1

0

(0.5)u

2

du =

1

6

.

u

v

1

v=(0.5)u 2

1

(b) P{Z 4} = P{Y 4X

2

} =

0.5

0

4u

2

du + 0.5 =

2

3

.

u

v

1

1 0.5

v=4u

2

3

- set1Uploaded bydee
- Maths - CHANCE & PROBABILITY Unit Plan Year 5/6Uploaded bymbed2010
- Stress Tests Into Mrket Risk ModellingUploaded byneuraltechsoft
- x2 x2-z = E = mc2 mv2 Einstein,Rossetti 1969. This opens the SFA file.Uploaded byMark Richard Hilbert(Mark Richard Rossetti)
- Eduachievers Chapter 30aUploaded byblue_l1
- Constant Potential Panels Slideshow (1)Uploaded byCarlos Gulbenkian Galdames
- Same La All OnlyUploaded byelaineliew_myun
- Coin Flip Independence Output probability stochasticsUploaded byrobin
- Eat 239 (Pipe Network) Example 4 SolutionUploaded bykhai9000
- HW2Uploaded byHayden Chappelear-Robbins
- stat11t_Chapter4 editmeUploaded byMuhammad Anshar
- Exercises Ch 6Uploaded byYazanAlomari
- V1I9-IJERTV1IS9379Uploaded bySuresh
- Bayes TheoremUploaded byAakash Kishore
- 2.1 ExampleUploaded byLucille Bautista
- ch02Uploaded byMardonio Jr Mangaser Agustin
- RIP Call for Paper Text PageUploaded bySuresh Kumar
- Chapter 15 More About ProbabilityUploaded byapi-3704862
- AssetUploaded byIvan Clark
- wawanUploaded byAdeAdhyaksa
- Unit 1 ( PROBABILITY THEORY ) - statistikUploaded byZara Nabilah
- lowcredit.pdfUploaded byAnonymous jiembd
- Risk Ranking ToolUploaded byKatia Matamoros Osorio
- Probability Models for Battle Damage AssessmentUploaded byAbraxas Luchenko
- Ch 5 HW SolutionsUploaded byanelesquivel
- Dougthery Social EffectsUploaded byIsaias Morales
- ProbabilityUploaded bySherifa
- Probability QsUploaded byKaran Tekwani
- WEB QUALIS - ENGENHARIAS IIIUploaded bydiogoscv
- Solution 01Uploaded byRahul Menon

- sol3Uploaded byThinh
- sol7Uploaded byThinh
- sol6Uploaded byThinh
- sol5Uploaded byThinh
- sol2Uploaded byThinh
- FinalUploaded byThinh
- sol4Uploaded byThinh
- Quiz SolUploaded byThinh
- exam2solUploaded byThinh
- QuizUploaded byThinh
- ps7Uploaded byThinh
- ps6Uploaded byThinh
- ps5Uploaded byThinh
- ps4Uploaded byThinh
- ps3Uploaded byThinh
- ps2Uploaded byThinh
- ps1Uploaded byThinh
- Final SolUploaded byThinh
- exam2Uploaded byThinh
- exam1Uploaded byThinh
- exam1solUploaded byThinh
- 534 CoversUploaded byThinh
- Minimax OptimizationUploaded byThinh

- Summary Applied Econometric Time SeriesUploaded byRashidAli
- Exercises for Stochastic CalculusUploaded byA.Benhari
- Logistic RegressionUploaded byJaya Verma
- bayesclustUploaded byAshwini Kumar Pal
- Conducting and Interpreting Canonical Correlation Analysis in Personality ResearchUploaded byElder Sobrinho
- SI417-Tut8-2017Uploaded byarjunvenugopalachary
- Tabel PoissonUploaded byKazama Ren
- Determining Sample SizeUploaded byعبدالفتاح البكوش
- CA Foun Nov. 2018 Mock Test BMRS (28.8.2018)Uploaded bypratik keshri
- GB MockexamUploaded byAnonymous 7K9pzvzi
- ABC.pdfUploaded byRenata Perez
- CRD RBD Factorial FRBD Split SPD Design AnalysisUploaded bySangita
- Retrial Queueing Model With Non Persistent Customers, Random Break Down, Delaying Repair and Bernoulli VacationUploaded byAlexander Decker
- R_lccUploaded byuit5
- Fisher, R. a. (1925) Statistical Methods for Research WorkersUploaded byberorakun
- RVS Certificate in Advanced Business Analytics and Data Science Using SAS and RUploaded bySrikanth Balasubramanian
- Planque and Fredou 1999.pdfUploaded byThierryF
- Descriptive Measures for Grouped DataUploaded byMikee Melad
- Fabriga, R. - Budget of Work (Math)Uploaded byReygie Logronio Fabriga
- Inference for RegressionUploaded byb3nzy
- Binary Logistic Regression Mintab TutorialUploaded byMuhammad Imdadullah
- Review Chapter 3Uploaded bynikowawa
- normality skewness kurtosisUploaded byapi-285777244
- Harvard Government 2000 SyllabusUploaded byJ
- Predicting VolatilityUploaded bygmustafaijt
- Exam 2 Practice QuestionsUploaded byvuduyduc
- MB0040 Statistics for Management Sem 1 Aug Spring AssignmentUploaded byNeelam Aswal
- CorrelationUploaded byhamzah
- MixtureUploaded byDaniloFreire
- cheatsheet statsUploaded bySebastian Kosch

## Much more than documents.

Discover everything Scribd has to offer, including books and audiobooks from major publishers.

Cancel anytime.