Journal of Statistical Software
November 2007, Volume 21, Issue 11.
boa: An R Package for MCMC Output Convergence
Assessment and Posterior Inference
Brian J. Smith
The University of Iowa
Markov chain Monte Carlo (MCMC
1. Generate a bivariate normal distribution with mean vector (
, and correlation , using Gibbs sampling.
2. Consider the following joint distribution
Lets say our goal is
1. Generate 50 random numbers based on the multiplicative congruential generator
method, and choose initial seed as
2. Generate 50 random numbers based on the linear congruential generator method,
and choose initial seed
ST430 Spring 2002
May 09, 2001, 11:20am1:20pm
There are ten questions, with a total of 110 points (but 100 is the full!). Please start a new page for each
question and do not use both sides. Lastly, enjoy the summer!
1. [Total 15 points]