fitted.mix.r

这个包里面含有利用极大似然估计方法来拟合成组数据和条件数据的有限混合分布模型的函数

## last modified May 2008

fitted.mix <- function(object, digits = NULL, ...) 
{
    mixobj<-object
    pmat <- grpintprob(mixobj$mixdata, mixobj$parameters, mixobj$distribution, 
        mixobj$constraint)
    n <- sum(mixobj$mixdata[, 2])
    joint <- n * sweep(pmat, 2, mixobj$parameters[, 1], "*")
    mixed <- apply(joint, 1, sum)
    conditprob <- sweep(joint, 1, mixed, "/")
    if (mixobj$usecondit) {
        conditional <- sweep(conditprob, 1, apply(mixobj$mixdata[, 
            -(1:2)], 1, sum), "*")
        outlist <- list(mixed = mixed, joint = joint, conditional = conditional, 
            conditprob = conditprob)
    }
    else outlist <- list(mixed = mixed, joint = joint, conditprob = conditprob)
    if (is.null(digits)) 
        outlist
    else sapply(outlist, round, digits)
}