R/ogsrl.r

In lrmest: Different Types of Estimators to Deal with Multicollinearity

ogsrliu<-function (formula, r, R, dpn, delt, d, data = NULL, na.action, 
    ...) 
{
    d <- as.matrix(d)
    d1 <- d[1L]
    ogsrles <- function(formula, r, R, dpn, delt, d1, data = NULL, 
        na.action, ...) {
        cal <- match.call(expand.dots = FALSE)
        mat <- match(c("formula", "data", "na.action"), names(cal))
        cal <- cal[c(1L, mat)]
        cal[[1L]] <- as.name("model.frame")
        cal <- eval(cal)
        y <- model.response(cal)
        md <- attr(cal, "terms")
        x <- model.matrix(md, cal, contrasts)
        s <- t(x) %*% x
        xin <- solve(s)
        r <- as.matrix(r)
        de1t <- as.matrix(delt)
        del <- delt
        RC <- matrix(R, NCOL(x))
        RR <- t(RC)
        I <- diag(NCOL(x))
        if (is.matrix(R)) 
            RR <- R
        else RR <- RR
        if (length(dpn) == 1L) 
            shi <- dpn
        else if (is.matrix(dpn)) 
            shi <- dpn
        else shi <- diag(dpn)
        bb <- xin %*% t(x) %*% y
        ev <- (t(y) %*% y - t(bb) %*% t(x) %*% y)/(NROW(x) - 
            NCOL(x))
        ev <- diag(ev)
        w1 <- solve(s/ev + t(RR) %*% solve(shi) %*% RR)
        w2 <- (t(x) %*% y)/ev + t(RR) %*% solve(shi) %*% r
        bm <- w1 %*% w2
        fd <- solve(s + I) %*% (s + d1 * I)
        srl <- fd %*% bm
srlve<-as.vector(srl)
j<-0
sumsq<-0
for (j in 1:NROW(srlve))
{
sumsq=(srlve[j])^2+sumsq
}
cval<-sumsq
ahat<-srl%*%t(srl)%*%solve(ev*xin+srl%*%t(srl))
ogsrle<-ahat%*%bb
        colnames(ogsrle) <- c("Estimate")
        dbd <- ev*(ahat%*%xin%*%t(ahat))
        Standard_error <- sqrt(diag(abs(dbd)))
        dbd <- ev*(ahat%*%xin%*%t(ahat))
rval<-(1/cval)*srl%*%t(srl)
mse1 <- cval^2*ev*tr(ev*rval*solve(ev*xin+cval*rval)%*%xin%*%solve(ev*xin+cval*rval)%*%rval)+ev^2*t(srl)%*%solve(ev*I+cval*rval%*%s)%*%solve(ev*I+cval*rval%*%s)%*%srl
mse1<-as.vector(mse1)       
rdel <- matrix(delt, NROW(RR))
        lenr <- length(RR)
        dlpt <- diag(RR %*% xin %*% t(RR))
        if (lenr == ncol(RR)) 
            ilpt <- sqrt(solve(abs(dlpt)))
        else ilpt <- sqrt(solve(diag(abs(dlpt))))
        upt <- RR %*% ogsrle
        tb <- t(upt)
        t_statistic <- ((tb - t(rdel)) %*% ilpt)/sqrt(ev)
        tst <- t(2L * pt(-abs(t_statistic), df <- (NROW(x) - 
            NCOL(x))))
        pvalue <- c(tst, rep(NA, (NCOL(x) - NROW(RR))))
        mse1 <- round(mse1, digits <- 4L)
        names(mse1) <- c("MSE")
        t_statistic <- c(t_statistic, rep(NA, (NCOL(x) - NROW(RR))))
        ans1 <- cbind(ogsrle, Standard_error, t_statistic, pvalue)
        ans <- round(ans1, digits <- 4L)
        anw <- list(`*****Stochastic Restricted Liu Estimator*****` = ans, 
            `*****Mean square error value*****` = mse1)
        return(anw)
    }
    npt <- ogsrles(formula, r, R, dpn, delt, d1, data, na.action)
    plotogsrliu <- function(formula, r, R, dpn, delt, d, data = NULL, 
        na.action, ...) {
        i <- 0
        arr <- 0
        for (i in 1:NROW(d)) {
            if (d[i] < 0L) 
                d[i] <- 0L
            else d[i] <- d[i]
            if (d[i] > 1L) 
                d[i] <- 1L
            else d[i] <- d[i]
            ogsrlm <- function(formula, r, R, dpn, del, d, data, 
                na.action, ...) {
                cal <- match.call(expand.dots = FALSE)
                mat <- match(c("formula", "data", "na.action"), 
                  names(cal))
                cal <- cal[c(1L, mat)]
                cal[[1L]] <- as.name("model.frame")
                cal <- eval(cal)
                y <- model.response(cal)
                md <- attr(cal, "terms")
                x <- model.matrix(md, cal, contrasts)
                s <- t(x) %*% x
                xin <- solve(s)
                r <- as.matrix(r)
                de1t <- as.matrix(delt)
                del <- delt
                RC <- matrix(R, NCOL(x))
                RR <- t(RC)
                I <- diag(NCOL(x))
                if (is.matrix(R)) 
                  RR <- R
                else RR <- RR
                if (length(dpn) == 1L) 
                  shi <- dpn
                else if (is.matrix(dpn)) 
                  shi <- dpn
                else shi <- diag(dpn)
                bb <- xin %*% t(x) %*% y
                ev <- (t(y) %*% y - t(bb) %*% t(x) %*% y)/(NROW(x) - 
                  NCOL(x))
                ev <- diag(ev)
w1 <- solve(s/ev + t(RR) %*% solve(shi) %*% RR)
        w2 <- (t(x) %*% y)/ev + t(RR) %*% solve(shi) %*% r
        bm <- w1 %*% w2
        fd <- solve(s + I) %*% (s + d * I)
        srl <- fd %*% bm
srlve<-as.vector(srl)
j<-0
sumsq<-0
for (j in 1:NROW(srlve))
{
sumsq=(srlve[j])^2+sumsq
}
cval<-sumsq
ahat<-srl%*%t(srl)%*%solve(ev*xin+srl%*%t(srl))
dbd <-ev*(ahat%*%xin%*%t(ahat))
rval<-(1/cval)*srl%*%t(srl)
mse1 <-cval^2*ev*tr(ev*rval*solve(ev*xin+cval*rval)%*%xin%*%solve(ev*xin+cval*rval)%*%rval)+ev^2*t(srl)%*%solve(ev*I+cval*rval%*%s)%*%solve(ev*I+cval*rval%*%s)%*%srl 
mse1<-as.vector(mse1)
                return(mse1)
            }
            arr[i] <- ogsrlm(formula, r, R, dpn, delt, d[i], data, 
                na.action)
        }
        MSE <- arr
        parameter <- d
        pvl <- cbind(parameter, MSE)
        colnames(pvl) <- c("Parameter", "MSE")
        sval <- pvl
        return(sval)
    }
    psrle <- plotogsrliu(formula, r, R, dpn, delt, d, data, na.action)
    if (nrow(d) > 1L) 
        val <- psrle
    else val <- npt
    val
}