R/fuzcluster.R

In sirt: Supplementary Item Response Theory Models

## File Name: fuzcluster.R
## File Version: 0.163


#*** Clustering for continuous fuzzy data
fuzcluster <- function(dat_m, dat_s, K=2, nstarts=7,
            seed=NULL, maxiter=100, parmconv=.001, fac.oldxsi=0.75, progress=TRUE )
{
    s1 <- Sys.time()
    dev0 <- 1e200
    dat_resp <- 1 - is.na(dat_m)
    if ( ! is.null(seed) ){
        nstarts <- 1
    }
    for ( rr in 1:nstarts ){
        res1 <- fuzcluster_estimate(K, dat_m, dat_s, dat_resp,
                    maxiter=maxiter, parmconv=parmconv, progress=progress,
                    seed=seed, fac.oldxsi=fac.oldxsi)
        if ( res1$deviance < dev0 ){
            res <- res1
            dev0 <- res1$deviance
        }
    }
    s2 <- Sys.time()
    ### end random starts
    ### computation of information criteria
    dev <- res$deviance
    ic <- list( deviance=dev, n=nrow(dat_m) )
    I <- ncol(dat_m)
    ic$np <- (K-1) + 2*K*I
    ic$AIC <- dev + 2*ic$np
    ic$BIC <- dev + ( log(ic$n) )*ic$np
    ic$CAIC <- dev + ( log(ic$n) + 1 )*ic$np
    ic$AICc <- ic$AIC + 2*ic$np * ( ic$np + 1 ) / ( ic$n - ic$np - 1 )
    res$ic <- ic
    res$K <- K
    res$s1 <- s1
    res$s2 <- s2
    res$nstarts <- nstarts
    class(res) <- "fuzcluster"
    return(res)
}