A prior guess at the value of the standard deviation.
sample.size
The weight given to sigma.guess.
Interpretable as a prior observation count.
initial.value
The initial value of the paramter in the MCMC
algorithm.
fixed
Logical. Some algorithms allow you to fix sigma at a
particular value. If TRUE then sigma will remain fixed at
initial.value, if supported.
upper.limit
If positive, this is the upper limit on possible
values of the standard deviation parameter. Otherwise the upper
limit is assumed infinite. Not supported by all MCMC algorithms.
References
Gelman, Carlin, Stern, Rubin (2003), "Bayesian Data Analysis", Chapman
and Hall.