R/eapmulc2.R

In InDisc: Obtaining and Estimating Unidimensional and Multidimensional IRT Dual Models

eapmulc2 <- function (mu, RES, lam, vres, nodos1, nodos2){

  RES<-as.matrix(RES)

  n<-size(RES)[1]
  m<-size(RES)[2]

  if (n>m){
    ni <- n
  }
  else {
    ni <- m
  }

  k <- dim(nodos1)[1]

  th <- matrix(0,1,2)
  vi <- 0
  st <- matrix(0,1,2)
  sv <- 0

  p <- matrix(0,ni)
  L1 <- array(0,dim=c(k,k,k))
  deno <- 0
  nume1 <- 0
  nume2 <- 0
  nume3 <- 0
  nume4 <- 0
  nume5 <- 0
  nume6 <- 0


  for (l in 1:k){
    for (t in 1:k){
      for (i in 1:k){

        L1[l,t,i] <- 1

        for (j in 1:ni){

          lamtmp <- cbind(lam[j,1],lam[j,2])
          rlamtmp <- sqrt(lamtmp %*% (transpose(lamtmp)))

          term0 <- rlamtmp %*% sqrt(nodos2[i,1]+vres[j])
          term1 <- 0.399 / term0
          term2 <- RES[j] - mu[j] - (lam[j,1] %*% nodos1[l,1]) - (lam[j,2] %*% nodos1[t,1])
          term3 <- (term2 / term0) %*% (term2 / term0)
          term4 <- exp(-0.5 * term3)
          term5 <- term1 * term4
          L1[l,t,i] <- L1[l,t,i] * term5
        }

        p <- as.matrix(L1[l,t,i] * nodos1[l,2] * nodos1[t,2] * nodos2[i,2]) # evitar named num

        deno <- deno + p
        nume1 <- nume1 + p * nodos1[l,1]
        nume2 <- nume2 + p * nodos1[t,1]
        nume3 <- nume3 + p * nodos2[i,1]

        nume4 <- nume4 + p * nodos1[l,1] * nodos1[l,1]
        nume5 <- nume5 + p * nodos1[t,1] * nodos1[t,1]
        nume6 <- nume6 + p * nodos2[i,1] * nodos2[i,1]
      }
    }
  }

  th[1] <- nume1 / deno
  th[2] <- nume2 / deno
  vi <- nume3 / deno

  st[1] <- sqrt(nume4 / deno - (th[1]*th[1]))
  st[2] <- sqrt(nume5 / deno - (th[2]*th[2]))
  sv <- sqrt(nume6 / deno - (vi*vi))

  # reli th and PDD
  reli_th <- 1 - (st^2)
  reli_PDD <- 0.4159/ (0.4159 + sv^2) # nodes variance

  OUT <- list("th"=th, "vi"=vi, "st"=st, "sv"=sv, "reli_th" = reli_th, "reli_PDD" = reli_PDD)

  return(OUT)

}