160a09_hw3 - explanation for your result. Problem 3 Problem...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Pstat160a Winter 2009 Hw #3 Problem 1 Let ( X,Y ) be discrete random variables with joint probability generating function G X,Y ( s 1 ,s 2 ) = exp ± λ 1 ( s 1 - 1) + λ 2 ( s 2 - 1) + γ ( ( s 1 - 1)( s 2 - 1)( s 1 - s 2 ) Show, using a general result from a previous discussion section, that X and Y are both Poisson random variable as well as X + Y . Identify the parameters for each case. Argue that this distribution provides a counter example showing that the converse of the property for sum of independent Poisson random variable shown in class is not true. Problem 2 Use the rejection method to simulate an arbitrary geometric random variable from a a geometric (1/2). For what values values of p would the procedure work? What is the expected number of iteration and give an intuitive
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: explanation for your result. Problem 3 Problem 14, ch 11, only a) b) and c). Problem 4 a) Let X be any random variable, and a R xed. Show that cov( X,a ) = 0. b) Let U U (0 , 1), and let V = 1-U . Show that cov( U,V ) < 0. Problem 5 Download the Matlab program rejectgeo.m from the class webpage, which implement the algorithm for problem 2 above. Explain it, run it for various values of p , check that indeed only some values of p work and comment. Explain what happens for the p that are not valid. Try also smaller and larger values for C (the bound for the ratio of pdf)....
View Full Document

This document was uploaded on 04/29/2009.

Ask a homework question - tutors are online