R/gdina_calc_expected_counts.R

In alexanderrobitzsch/CDM: Cognitive Diagnosis Modeling

## File Name: gdina_calc_expected_counts.R
## File Version: 0.152

gdina_calc_expected_counts <- function( G, J, L, item.patt.freq, p.aj.xi, some.missings,
        ipr, attr.patt.c, resp.patt, item.patt.split, data )
{
    R.lj.gg <- I.lj.gg <- NULL
    #---------------------- single groups ------------------------------------------
    if (G==1){
        if ( some.missings ){
            I.lj <- crossprod( item.patt.freq*resp.patt    , p.aj.xi )
        } else {
            I.lj <- matrix( crossprod( item.patt.freq, p.aj.xi ), nrow=J, ncol=L,
                                            byrow=TRUE )
        }
        R.lj <- crossprod(ipr, p.aj.xi )
        colnames(I.lj) <- colnames(R.lj) <- attr.patt.c
        rownames(I.lj) <- rownames(R.lj) <- colnames(data)
    }    # end one group

    #---------------------- multiple groups ------------------------------------------
    if (G > 1){
        R.lj.gg <- I.lj.gg <- array( 0, c( J, L, G ) )
        for (gg in 1L:G){
            I.lj.gg[,,gg] <- crossprod( item.patt.freq[,gg]*resp.patt, p.aj.xi[,,gg] )
            R.lj.gg[,,gg] <- crossprod( item.patt.split  * item.patt.freq[,gg] *
                                            resp.patt, p.aj.xi[,,gg] )
            colnames(I.lj.gg) <- colnames(R.lj.gg) <- attr.patt.c
            rownames(I.lj.gg) <- rownames(R.lj.gg) <- colnames(data)
        }
        # calculate I.lj and R.lj
        I.lj <- I.lj.gg[,,1]
        R.lj <- R.lj.gg[,,1]
        for (gg in 2L:G){
            I.lj <- I.lj + I.lj.gg[,,gg]
            R.lj <- R.lj + R.lj.gg[,,gg]
        }
    }
    #--- OUTPUT
    res <- list( I.lj=I.lj, R.lj=R.lj, I.lj.gg=I.lj.gg, R.lj.gg=R.lj.gg)
    return(res)
}