Skip to content
Prev 168638 / 398502 Next

gls prediction using the correlation structure in nlme

On Wed, Jan 28, 2009 at 9:51 AM, Dr Carbon <drcarbon at gmail.com> wrote:
I think this is the way to do this?
b0 <-  coef(fm1)[1]
b1 <- coef(fm1)[2]
p1 <- intervals(fm1)$corStruct[2]
y[i] = b0 + p1*y[i-1] + b1*x[i]