Utah State University: ECE 6010 Homework 4 - Problems on Stochastic Processes, Assignments of Stochastic Processes

Homework problems for the utah state university electrical and computer engineering (ece) 6010 course on stochastic processes. The problems cover topics such as mean and covariance of normal distributions, generating random vectors from normal distributions, minimum variance unbiased estimation, and multivariate normal distributions.

Typology: Assignments

Pre 2010

Uploaded on 07/30/2009

koofers-user-78i
koofers-user-78i 🇺🇸

3.3

(3)

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Utah State University
ECE 6010
Stochastic Processes
Homework # 4
Due Friday Oct. 7, 2005
1. Suppose X N (µ,Σ).
(a) Show that E[X] = µand cov(X,X) = Σ.
(b) Show that AX+b N (Aµ+b, AΣAT).
(c) Suppose Σ >0 and write Σ = C CT. Show that C1(Xµ) N (0, I).
2. Suppose you have a random number generator which is capable of generating random
numbers distributed as X N (0,1). Describe how to generate random vectors Y
N(µ,Σ).
3. Suppose Xand Yare r.v.s. Show that E[(Xh(Y))2] is minimized over all functions
hwhen his the function
h(y) = E[X|Y=y].
Assume E[X2]<.
4. Let (X, Y ) N (µx, µy, σ 2
x, σ2
y, ρ). Let X= [ X
Y] N(µ,Σ), where
Σ = s11 s12
s21 s22.
Determine the relationship between µx, µy, σ2
x, σ2
y, ρ and µand Σ.
5. Suppose X=hX1
X2
X3i N(µ,Σ) where
µ=
1
2
3
Σ =
4 2 1
2 6 3
1 3 8
(a) The value X1= 1.5 is measured. Determine the best estimate for (X2, X3).
(b) In a separate problem, the values X2= 1 and X3= 5 are measured. Determine
the best estimate of X1.
1
pf2

Partial preview of the text

Download Utah State University: ECE 6010 Homework 4 - Problems on Stochastic Processes and more Assignments Stochastic Processes in PDF only on Docsity!

Utah State University

ECE 6010

Stochastic Processes

Homework # 4

Due Friday Oct. 7, 2005

  1. Suppose X ∼ N (μ, Σ). (a) Show that E[X] = μ and cov(X, X) = Σ. (b) Show that AX + b ∼ N (Aμ + b, AΣAT^ ). (c) Suppose Σ > 0 and write Σ = CCT^. Show that C−^1 (X − μ) ∼ N ( 0 , I).
  2. Suppose you have a random number generator which is capable of generating random numbers distributed as X ∼ N (0, 1). Describe how to generate random vectors Y ∼ N (μ, Σ).
  3. Suppose X and Y are r.v.s. Show that E[(X − h(Y ))^2 ] is minimized over all functions h when h is the function h(y) = E[X|Y = y]. Assume E[X^2 ] < ∞.
  4. Let (X, Y ) ∼ N (μx, μy, σ x^2 , σ^2 y , ρ). Let X = [ XY ] ∼ N (μ, Σ), where

Σ =

[s 11 s 12 s 21 s 22

]

Determine the relationship between μx, μy, σ^2 x, σ y^2 , ρ and μ and Σ.

  1. Suppose X =

[ X 1

X X (^23)

]

∼ N (μ, Σ) where

μ =

(a) The value X 1 = 1.5 is measured. Determine the best estimate for (X 2 , X 3 ). (b) In a separate problem, the values X 2 = 1 and X 3 = 5 are measured. Determine the best estimate of X 1. 1

(c) Determine a random vector Y which is a whitened version of X.

Problems from Grimmet & Stirzaker

  1. Ex 3.7.5. What is requested is E[T − t|T > t], i.e., the mean subsequent lifetime given that the machine is still running after t days. Then use the hint from the book. Note that in (a), P (T > t) = (^) N^1 +1 (N − t).
  2. Ex 3.7.7. Hint: Show that P (robot faulty|fault not detected) = φ 1 (1−−φδδ) =^4 π. Hence argue that the number of faulty passed robots, given Y , is distributed as B(n − Y, π), which has mean (n − Y )π. Hence show that E[X|Y ] = Y + (n − Y )π.
  3. Ex 4.1.1(a)
  4. Ex 4.1.
  5. Ex 4.2.1. Hint: think geometric r.v.
  6. Ex 4.2.2. Hint: P (max(X, Y ) ≤ v) = P (X ≤ v, Y ≤ v).
  7. Ex 4.4.1. Hint: integrate by parts.
  8. Ex 4.6.4(b)