Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

1. Let XX have density fX(x)=2xf_X(x) = 2x for 0<x<10 < x < 1. Find the density of

(a) 5X35X - 3

(b) 4X34X^3

(c) X/(1+X)X/(1+X)

2. Let XX have density f(x)=2x3ex2f(x) = 2x^{-3}e^{-x^{-2}} on the positive real numbers. Find the density of X4X^4.

3. For a fixed α>0\alpha > 0 let XX have the Pareto density given by

f(x) = αxα+1,  x>1f(x) ~ = ~ \frac{\alpha}{x^{\alpha+1}}, ~~ x > 1

Find the density of log(X)\log(X). Recognize this as one of the famous ones and provide its name and parameters.

4. Let XX be a random variable. Find the density of X2X^2 if XX has the uniform distribution on (1,2)(-1, 2).

5. Let UU have the uniform (0,1)(0, 1) distribution. For λ>0\lambda > 0, find a function of UU that has the exponential (λ)(\lambda) distribution.

6. Let ZZ be standard normal.

(a) Use the change of variable formula to find the density of 1/Z1/Z. Why do you not have to worry about the event Z=0Z = 0?

(b) A student who doesn’t like the change of variable formula decides to first find the cdf of 1/Z1/Z and then differentiate it to get the density. That’s a fine plan. The student starts out by writing P(1/Z<x)=P(1/x<Z)P(1/Z < x) = P(1/x < Z) and immediately the course staff say, “Are you sure?” What is the problem with what the student wrote?

(c) For all xx, find P(1/Z<x)P(1/Z < x).

(d) Check by differentiation that your answer to (c) is consistent with your answer to (a).

7. Let the random variable XX have cdf FF and let the random variable YY have cdf GG. You can assume that both FF and GG are continuous and increasing.

(a) Find a function hh such that the random variable h(X)h(X) has the uniform (0,1)(0, 1) distribution.

(b) Use Part (a) to find a function gg such that the random variable g(X)g(X) has the same distribution as YY.