R/TransMatrix.R

In TPmsm: Estimation of Transition Probabilities in Multistate Models

TransMatrix.TPmsm <- function(x) {
	mat <- matrix(0, nrow=3, ncol=3);
	rownames(mat) <- colnames(mat) <- x$state.names;
	if ( !is.null(x$n.boot) ) {lst <- vector(mode="list", length=3);}
	else {lst <- vector(mode="list", length=1);}
	n <- length(x$time);
	for ( i in 1:length(lst) ) {
		mat[1,1] <- x[[i+1]][n,1];
		mat[1,2] <- x[[i+1]][n,2];
		mat[1,3] <- x[[i+1]][n,3];
		mat[2,2] <- x[[i+1]][n,4];
		mat[2,3] <- x[[i+1]][n,5];
		mat[3,3] <- 1;
		lst[[i]] <- mat;
	}
	return(lst);
} # TransMatrix.TPmsm

TransMatrix.TPCmsm <- function(x) {
	mat <- matrix(0, nrow=3, ncol=3);
	rownames(mat) <- colnames(mat) <- x$state.names;
	nx <- length(x$x);
	lst <- vector(mode="list", length=nx);
	if ( !is.null(x$n.boot) ) {nm <- 3;}
	else {nm <- 1;}
	nt <- length(x$time);
	for (i in 1:nx) {
		lst[[i]] <- vector(mode="list", length=nm);
		ix <- which(x$covariate == x$x[i]);
		for (j in 1:nm) {
			mat[1,1] <- x[[j+1]][nt, ix, 1];
			mat[1,2] <- x[[j+1]][nt, ix, 2];
			mat[1,3] <- x[[j+1]][nt, ix, 3];
			mat[2,2] <- x[[j+1]][nt, ix, 4];
			mat[2,3] <- x[[j+1]][nt, ix, 5];
			mat[3,3] <- 1;
			lst[[i]][[j]] <- mat;
		}
	}
	return(lst);
} # TransMatrix.TPCmsm