I have an unexpected result with the functions lag() and diff() in the plm
(panel data) package when used with transform(). These plm-specific
functions are supposed to generate lags and first differences within each
panel.
lag() does not work properly the first time (it reproduces the same
series--this is a common time series pitfall), BUT then it does work
properly when it is run a second time. (This is demonstrated in the code
below.
diff() does not work at all if it is run before lag(). It works properly if
it is run after lag(). The problem occurs only inside a transform
statement. lag() and diff() work correctly with Grunfeld.p$l.value <-
lag(Grunfeld.p$value,1)
Grunfeld.p$d.value <- diff(Grunfeld.p$value,1)
Here is code to reproduce the error:
library(plm)
data(Grunfeld)
Grunfeld.p <- pdata.frame(Grunfeld,c("firm","year"))
Grunfeld.p <- transform(Grunfeld.p, l.value = lag(value,1))
head(Grunfeld.p)
Grunfeld.p <- transform(Grunfeld.p, l.value = lag(value,1))
head(Grunfeld.p)
Run the following in a separate R session:
library(plm)
data(Grunfeld)
Grunfeld.p <- pdata.frame(Grunfeld,c("firm","year"))
Grunfeld.p <- transform(Grunfeld.p, d.value = diff(value,1))
head(Grunfeld.p)
Grunfeld.p <- transform(Grunfeld.p, l.value = lag(value,1))
Grunfeld.p <- transform(Grunfeld.p, d.value = diff(value,1))
head(Grunfeld.p)
R version 2.13.0 (2011-04-13)
Platform: i686-pc-linux-gnu (32-bit)
other attached packages:
[1] plm_1.2-6 sandwich_2.2-6 zoo_1.6-4 MASS_7.3-13
[5] Formula_1.0-0 kinship_1.1.0-23 lattice_0.19-26 nlme_3.1-101
[9] survival_2.36-9
Information on package 'plm'
Description:
Package: plm
Version: 1.2-6
Date: 2010-11-20
Title: Linear Models for Panel Data
--
Michael Ash, Associate Professor
of Economics and Public Policy
Department of Economics and CPPA
University of Massachusetts
Amherst, MA 01003
Email m...@econs.umass.edu
Tel +1-413-545-6329 // Fax +1-413-545-2921
http://people.umass.edu/maash
--
Michael Ash, Associate Professor
of Economics and Public Policy
Department of Economics and CPPA
University of Massachusetts
Amherst, MA 01003
Email m...@econs.umass.edu
Tel +1-413-545-6329 // Fax +1-413-545-2921
http://people.umass.edu/maash
[[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.
