The NormalInverseGammaPrior is the conjugate prior for the mean and variance of the scalar normal distribution. The model says that $$\frac{1}{\sigma^2} \sim Gamma(df / 2, ss/2) \mu|\sigma \sim N(\mu_0, \sigma^2/\kappa)$$
NormalInverseGammaPrior(mu.guess, mu.guess.weight = .01,
sigma.guess, sigma.guess.weight = 1, ...)The mean of the prior distribution. This is \(\mu_0\) in the description above.
The number of observations worth of weight
assigned to mu.guess. This is \(\kappa\) in the
description above.
A prior estimate at the value of sigma.
This is \(\sqrt{ss/df}\).
The number of observations worth of weight
assigned to sigma.guess. This is \(df\).
blah
Steven L. Scott steve.the.bayesian@gmail.com
Gelman, Carlin, Stern, Rubin (2003), "Bayesian Data Analysis", Chapman and Hall.