est_prior: Fit the mixture inverse-gamma prior of variance
In mengyin/vashr: Variance Adaptive Shrinkage

Fit the mixture inverse-gamma prior of variance, given the variance estimates (sehat^2).

1 2	est_prior(sehat, df, betahat, randomstart, singlecomp, unimodal, prior, g, maxiter, estpriormode, priormode, completeobs)

`sehat`	a p vector of observed standard errors
`df`	appropriate degrees of freedom for (chi-square) distribution of sehat
`betahat`	a p vector of estimates (optional)
`randomstart`	logical, indicating whether to initialize EM randomly. If FALSE, then initializes to prior mean (for EM algorithm) or prior (for VBEM)
`singlecomp`	logical, indicating whether to use a single inverse-gamma distribution as the prior distribution for the variances
`unimodal`	put unimodal constraint on the prior distribution of variances ("variance") or precisions ("precision")
`prior`	string, or numeric vector indicating Dirichlet prior on mixture proportions (defaults to "uniform", or 1,1...,1; also can be "nullbiased" 1,1/k-1,...,1/k-1 to put more weight on first component)
`g`	the prior distribution for variances (usually estimated from the data; this is used primarily in simulated data to do computations with the "true" g)
`maxiter`	maximum number of iterations of the EM algorithm
`estpriormode`	logical, indicating whether to estimate the mode of the unimodal prior
`priormode`	specified prior mode (only works when estpriormode=FALSE).
`completeobs`	a p vector of non-missing flags