R/LBIC.R

In fchamroukhi/HDME: LASSO Regularized Mixture of Experts Models

#BIC  function
LBIC = function(X, Y, wk, eta)
{
  # source("Pik.R")
  # source("LPi.R")
  d = dim(X)[2]
  p = d-1
  n = dim(X)[1]
  K = nrow(wk) + 1
  R = max(Y)
  pik = Pik(n, K, X, wk)
  # beta = as.matrix(betak[-1,]) #remove first row of betak
  # if(p==1) beta = t(beta)
  w = wk[,-1] #remove first column of wk
  DF = 0
  for(k in 1:K)
  {
    for (r in 1:(R-1))
    {
      for (j in 2:d)
        if(eta[k,r,j]!= 0) DF = DF+1
    }
  }
  if(K==2) w = t(as.matrix(w))
  for(k in 1:(K-1))
  {
    for (j in 1:p)
      if(w[k,j]!= 0) DF = DF+1
  }
  S0 = 0
  for(i in 1:n)
  {
    S1 = 0
    for(k in 1:K)
    {
      ETAk = as.matrix(eta[k,,])
      if(R==2) ETAk = t(ETAk)
      P_eta = Pi(R,X[i,],ETAk)
      S1 = S1+pik[i,k]*P_eta[Y[i]]
    }
    S1 = log(S1)
    S0 = S0+S1
  }
  S0 = S0 - log(n)*DF/2
  return (S0)
}