R/gdm_calc_counts.R

In alexanderrobitzsch/CDM: Cognitive Diagnosis Modeling

## File Name: gdm_calc_counts.R
## File Version: 0.09

################################################
# calculation of expected counts
gdm_calc_counts <- function(G, weights, dat.ind, dat, dat.resp,
            p.aj.xi, K, n.ik, TP,I,group, dat.ind2, ind.group,
            use.freqpatt )
{
    # n.ik [ 1:TP, 1:I, 1:(K+1), 1:G ]
    # N.ik [ 1:TP, 1:I,  1:G ]
    N.ik <- array( 0, dim=c(TP,I,G) )
    if (G==1){
        gg <- 1
        for (kk in 1:(K+1)){   #        kk <- 1    # category 0 ( -> 1 )
            dkk2 <- weights*dat.ind2[[kk]][[gg]]
            n.ik[,,kk,gg] <- crossprod( p.aj.xi, dkk2 )
            N.ik[,,gg] <- N.ik[,,gg] + n.ik[,,kk,gg]
        }
    }
    if (G>1){
        for (gg in 1:G){    # gg <- 1
            ind.gg <- ind.group[[gg]]
            if ( ! use.freqpatt ){
                t.p.aj.xi.gg <- t( p.aj.xi[ind.gg,] )
            }
            if ( use.freqpatt ){
                t.p.aj.xi.gg <- t( p.aj.xi[[gg]] )
            }
            for (kk in 1:(K+1)){   #        kk <- 1    # category 0 ( -> 1 )
                dkk2 <- weights[ind.gg] * dat.ind2[[kk]][[gg]]
                if ( use.freqpatt ){
                    if (G>1){
                        dkk2 <- dkk2[ which(weights[,gg] > 0), ]
                    }
                }
                n.ik[,,kk,gg] <- t.p.aj.xi.gg %*% dkk2
                N.ik[,,gg] <- N.ik[,,gg] + n.ik[,,kk,gg]
            }
        }
    }
    #--- OUTPUT
    res <- list("n.ik"=n.ik, "N.ik"=N.ik )
    return(res)
}

.gdm.calc.counts <-    gdm_calc_counts