STAT 330 October 9, 2012

Here is a link to the next set of charts, on Poisson inference.

Here are links to the two code files I mentioned today, MCMC4.R and MCMC5.R. The problem with the first file was that a stray invisible character had somehow inserted itself into the file. I eliminated it and the code ran just fine. We’ll discuss this on Thursday.

Here is a link to a download of a ten-minute discussion with Nate Silver, the statistician who is using Bayesian methods to make forecasts of elections on his blog, fivethirtyeight. He discusses the basic ideas of what he’s doing.

Today we discussed more on MCMC simulation and looked at some of the things that have to be taken into account when programming a sampler using Metropolis-Hastings, Gibbs sampling, or hybrids such as Metropolis-Within-Gibbs. I then attempted to run the MCMC4.R file, which failed to run because as I later discovered it had acquired an invisible character somehow that compromised its integrity. I simply redid that bit of the file and now it runs fine. I did not change the code, that is, the surmise that we needed an explicit “return” statement was not true.

We then turned to Poisson inference, which describes situations in which events happen (stars, arrivals at the ER, radioactive decays, photon counting) at some rate, and we want to understand the counting process. I motivated the discussion by talking about filling a bowl with oatmeal that had some raisins in it. We briefly discussed priors and some other issues before class let out.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: