The dyn package (and also the dynlm package) can be used for
this.
library(dyn)
y <- ts(1:10)
dyn$lm(y ~ lag(y, -1))
On Mon, Jun 29, 2009 at 2:50 PM, Hongwei Dong wrote:
> Hi, I'm using R to do a time series analysis. In the model, I use the lags
> of some variables. such the lags of the variable
Like this?
x<-1:100
y<-runif(100)
df<-data.frame(cbind(x,y))
mycoef.list<-NULL
for (i in 10:65)
{
mycoef<-coef(lm(y~x, data=subset(df, x %in% 1:i)))[2]
mycoef.list <-c(mycoef.list, mycoef)
}
hist(mycoef.list)
regards
milton
brazil=toronto
On Mon, Jun 29, 2009 at 2:50 PM, Hongwei Dong wrot
Hi, I'm using R to do a time series analysis. In the model, I use the lags
of some variables. such the lags of the variables have different length, I
just can't use them directly in the lm function. Intuitively, I feel that
"subset" might be useful, but I do not know how to use it. Anyone can give
3 matches
Mail list logo
