Answer Session 1 Para 3.1

Here the prior was B(1,1) ( a flat prior) restricted to lie in the range (0.1,0.6)

An initial 100 samples were run, then pi was monitored nad 5000 samples stored.

Results from the STATS button give

Node mean sd MC error 2.5% median 97.5% start sample

pi 0.348 0.1433 0.001993 0.1129 0.3461 0.5876 1001 5000

So we have 5000 samples stored and a mean of 0.348.

The kernel density estimate gives