
June 3, 2023
	LS_empirical considers that beta ~ Student t
		but that se is fixed. 
		Can I draw sampling varinces from scaled chisq

	I currently use standard error, should I use shrunkedn standard error from eBayes??? backtransformign from the t shrinking?	


	Given V, sample from Wishart then draw beta from this



Oct 11, 2022
# TODO in mvTest 
# return number of features
# return stat1 and stat2 for RE2C
# return NA for stat1 if 1 features
# return per-component p-value


p <- 1
q <- 5
n <- 100000 # number of samples
# parameter specification
beta <- matrix(rnorm(p*q),p,q) # single argument

V = diag(3, q)

nu <- 10

# simulate n draws
res = rMNIW(n, Lambda = beta, Psi = Psi, nu = nu)


nu = rep(1e8, 5)

stats1 = get_stat_samples(V, nu, seed=1)
stats2 = get_stat_samples2(V, nu)

source("/Users/gabrielhoffman/workspace/repos/remaCor/R/LS_empirical_pvalue.R")
















