Regression Markov chains

In the previous exercise, you ran 4 parallel Markov chains of length 1,000 to approximate the posterior model of regression parameters \(a\), \(b\), and \(s\). Your output was similar to that below:

Though it requires too much computing power to run here, the results of a new RJAGS simulation run for 100,000 iterations is in your workspace. This mcmc.list object is stored as weight_sim_big. Construct and examine a plot() of the Markov chains in weight_sim_big. Which of the following best describes the comparison of these with the original simulation results?

Possible answers

The Markov chains of length 1,000 are too short. They have not stabilized, thus are unlikely to provide a reliable approximation of the posterior.

The Markov chains appear to stabilize after 100,000 iterations, thus provide a more reliable approximation of the posterior.

Both of the above.

Exercise

Regression Markov chains

Instructions

Possible answers