Let X and Y be two jointly Gaussian real random variables, each with zero mean,variance 1, and correlation coefficient ρ ∈ (0, 1). Let a, b ∈ R be such that a2 + b2 = 1, and defineW := aX + bY .a. Find values for a and b to maximize the variance of W . Hint: Use eigendecomposition.b. Does the optimal W (from part a) have a probability density function? If yes, derive it. Ifnot, explain why.

Let X and Y be two jointly Gaussian real random variables, each with zero mean,variance 1, and correlation coefficient ρ ∈ (0, 1). Let a, b ∈ R be such that a2 + b2 = 1, and defineW := aX + bY .a. Find values for a and b to maximize the variance of W . Hint: Use eigendecomposition.b. Does the optimal W (from part a) have a probability density function? If yes, derive it. Ifnot, explain why.

Linear Algebra: A Modern Introduction

4th Edition

ISBN:9781285463247

Author:David Poole

Publisher:David Poole

Chapter3: Matrices

Section3.7: Applications

Problem 13EQ

See similar textbooks

Similar questions

Find the moment-generating function of the contin-uous random variable X whose probability density is given by f(x) =1 for 0 < x < 10 elsewhere and use it to find μ1,μ2, and σ2.
Suppose that the random variables X,Y, and Z have the joint probability density function f(x,y,z) = 8xyz for 0<x<1, 0<y<1, and 0<z<1. Determine P(X<0.7).
If X and Y have the joint probability distributionf(−1, 0) = 0, f(−1, 1) = 1 4 , f(0, 0) = 16 , f(0, 1) = 0, f(1, 0) = 112 , and f(1, 1) = 12 , show that (a) cov(X, Y) = 0;(b) the two random variables are not independent.
Suppose a continuous random variable X has the following CDF:: F(x) = 1 - 1/ (x+1)4, x > 0.Find SX (x), survival function? a 1/(x+1)4, x < 0 b -1/(x+1)4, x > 0 c (x+1)4, x > 0 d 1/(x+1)4, x > 0
Let X be a standard normal random variable, X ∼ N(0, 1). Find a real-valued scalar k such that: E[(aX^2 + b)^k ] = 1 where E[·] denotes expectation, and a, b are positive real numbers satisfying a + b < 1.
(b) Let Z be a discrete random variable with E(Z) = 0. Does it necessarily follow that E(Z³) = 0? If yes, give a proof; if no, give a counterexample.
Consider a random variable, Y , which has a quasi-Bernoulli structure. With probability p ∈ [0, 1] it takes value 0. With probability (1 − p) it is described by a continuous random variable, X , with the following PDF, f_X (x)=1+x, x∈[−1,0), −1≤x<0f_X (x)=1−x, x∈[0,1], 0≤x≤1f_X (x)=0, otherwise , x>1 Obtain CDF of Y, F_Y (y), and draw the sketch.
Prove that for a continuous random variable X,E (aX+ b) = aE (X) + b.
Let X1 and X2 be two independent random variables. Suppose each Xi is exponentially distributed with parameter λi. Let Y=Min (X1, X2). A) Find the pdf of Y. B) Find E(Y). Hint: Let Y = Min (X1, X2). 1. P[Y > c] = P[Min (X1, X2) > c] = P[X1 > c, X2 > c] 2. Obtain the pdf of Y by differentiating its cdf of Y.
Let U1, ....U5 be independent and standard uniform distibuted random variables given by P(U1 ≤ x) = x, 0 < x < 1 1. Compute the moment generating function E(e sU ) of the random variable U1. 2. Compute the moment generating function of the random variable Y = aU1 + U2 + U3 + U4 + U5 with a > 0 unknown. 3. Compute E(Y ) and V ar(Y ). 4. As an estimator for the unknow value θ = a we migth use as an estimator θb = 2 n Xn i=1 Yi − 4 = 2Y − 4. with Yi independent and identically distributed having the same cdf as the random variable Y discussed in part 2. Compute E(θb) and V ar(θb) and explain why this estimator is sometimes not very useful. 5.Give an upperbound on the probability P(| θb− a |> ) for every > 0.(Hint:Use Chebyshevs inequality!)
Verify that p(x) = 3x - 4 is a probability density function on [1, oo)and calculate its mean value.
Find E(R) and V (R) for a random variable R whose moment-generating function ismR(t) = e2t(1-3t2)-1