EMAlgo {MixThres} | R Documentation |
EM algorithm to estimate the parameters of the mixture model.
EMAlgo(mdata, param, stop.crit)
mdata |
A vector of numeric. |
param |
A vector of parameters. |
stop.crit |
The stopping relative precision limit for stopping EM algorithm. |
The traditionnal EM algorithm of Depster et al. is modified in order to correct the mean and variance biaises du to the truncatures by including a fixed-point algorithm in the M-step. Then the vector epsilon algorithm is used to accelerate the convergence of the EM algorithm.
A list of result of the logLike function and a matrix of the posterior probabilities of belonging to each component of the mixture for each probe.
J. Aubert
Picard et al., Using mixture models to define a hybridization threshold in DNA microarray experiments.
Kuroda M., Sakakihara M., (2006), Accelerating the convergence of the EM algorithm using the vector epsilon algorithm, Computational Statistics & Data Analysis 51.