> var(as.vector(z.matrix)) [1] 0
I got sensible results from the following:
z.matrix <- outer(theta.1, theta.2, "+") > > ## Plot > contour(theta.1, theta.2, z.matrix)
hope this helps. spencer graves
Rafael Jimenez wrote:
Dear list,
I have been trying to use the function outer to make contour plots of a cost function for regression analysis with no success. (see commented code below).
Could anybody give me some advice on what I am doing wrong, or what is relevant for me to read?
Thanks, Rafael
## ============================================= CODE ## Create vectors and matrix Y <- c(0.5116403,2.4055245,1.6596707,1.8286057,2.6119199)
col.1 <- c(0.8997692,0.8216292,0.6449104,0.8179743,0.6602276) col.2 <- c(0.01175669,0.89389797,0.19913807,0.29872301,0.66144258)
X <- cbind(col.1, col.2)
## Define function ## Give default values so that it "takes two arguments" J <- function (x1, x2, X.val=X, Y.val=Y) { t(Y.val-X.val[,1]*x1-X.val[,2]*x2) %*% (Y.val-X.val[,1]*x1-X.val[,2]*x2) }
## Get grid to plot theta.1 <- array(seq(-1, 1, length=10)) theta.2 <- theta.1
## How to do this with outer? z.matrix <- outer(theta.1, theta.2, "J")
## Plot contour(theta.1, theta.2, z.matrix)
______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
