I have tried the diff and lag function.
I have given an example below. Diff gives me a$var2.shift.
I want a$var2.noshift, i.e. var^(t+2)-var^(t) should stay in the row with
year=t.
Diff shifts it to the t+2 row.
> a <- data.frame(c(rep(2,5), rep(3,5)), c(2005:2009, 2004:2008),
c(NA,10,34,23,12, 23,45, NA, 45, NA))
> colnames(a) <- c("firm","year","var")
> a <- pdata.frame(a)
> a$var2.shift <- diff(a$var, lag=2, difference=1)
> a$var2.noshift <- c(NA, 13,-22,NA, NA,NA, 0, NA, NA, NA)
> a
firm year var var2.shift var2.noshift
2-2005 2 2005 NA NA NA
2-2006 2 2006 10 NA 13
2-2007 2 2007 34 NA -22
2-2008 2 2008 23 13 NA
2-2009 2 2009 12 -22 NA
3-2004 3 2004 23 NA NA
3-2005 3 2005 45 NA 0
3-2006 3 2006 NA NA NA
3-2007 3 2007 45 0 NA
3-2008 3 2008 NA NA NA
On Tue, May 8, 2012 at 6:21 PM, Liviu Andronic <landronim...@gmail.com>wrote:
> On Tue, May 8, 2012 at 12:14 PM, Apoorva Gupta <apoorva.ni...@gmail.com>
> wrote:
> > I have checked that. It allows me to get the t-1, t-2 value but not the
> t+1
> > value.
> > Is there any other way of achieving this other than using the plm
> package?
> >
> It would be easier to help if you provided a minimal reproducible
> example, as requested in the posting guide. Have you tried
> diff(x, lag = -1, ...)
>
> or
> lag(x, k = -1, ...)
>
> Perhaps this does what you want. Regards
> Liviu
>
--
Apoorva Gupta
Consultant
National Institute of Public Finance and Policy
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
