Re: [R] Regress a matrix on another matrix column by column

2010-03-18 Thread Henrique Dallazuanna
Try this: mapply(function(y, x, z)lm(y ~ x + z), as.data.frame(y), as.data.frame(x.a), as.data.frame(x.b), SIMPLIFY = FALSE) On Thu, Mar 18, 2010 at 8:35 AM, Frederick Ho wrote: > Hi everyone, > > I have a response matrix (y) and two predictor matrices (x.a, x.b), how > should i proceed if i wan

Re: [R] Regress a matrix on another matrix column by column

2010-03-18 Thread Dimitris Rizopoulos
try this: y <- matrix(rnorm(100*100), 100, 100) x.a <- matrix(rnorm(100*100), 100, 100) x.b <- matrix(rnorm(100*100), 100, 100) M <- ncol(y) models <- vector("list", M) for (m in 1:M) { Dat <- data.frame(y = y[, m], x.a = x.a[, m], x.b = x.b[, m]) models[[m]] <- lm(y ~ ., data = Dat) }

[R] Regress a matrix on another matrix column by column

2010-03-18 Thread Frederick Ho
Hi everyone, I have a response matrix (y) and two predictor matrices (x.a, x.b), how should i proceed if i want to regress y on x.a and x.b column by column? To be specific, what i want to do is: y[,1]~x.a[,1]+x.b[,1] y[,2]~x.a[,2]+x.b[,2] . . . I have tried lm(y~x1+x2) but it does not work as