search for: mlogl_process

Displaying 1 result from an estimated 1 matches for "mlogl_process".

2008 May 23
0
Est. Component Size with AIC/BIC under Gamma Distribution
...puting log_likelihood return( - sum(dgamma(samp, shape = x_func, scale=theta_func, log = TRUE))) } find_bic <- function(mll,smpl,k) { bic <- (-2 * mll) + (k * log(length(smpl))) bic } find_aic <- function(mll,smpl,k) { aic <- (-2 * mll) + (k * 2) aic } mlogl_process <- function(smpl,error,start ) { # EM algorithm to estimate max loglikelihood thetalast <- 0 thetacurrent <- start theta <- start difference <- start thetaprocess <- start best <- 0 while (difference > error) { mlogl_out <- nlm(ml...