Dear Matt,
yes you have. 'plm' doesn't support multi-column time indices, but
it should happily make do with any *single* time index whose order can be
recognized by R, such as, e.g., a lexicographic ordering.
So you will probably want to paste your indices along these lines:
> year <- rep(2000:2001,times=c(4,4))
> year
[1] 2000 2000 2000 2000 2001 2001 2001 2001>
> quarter <- rep(1:4,times=2)
> quarter
[1] 1 2 3 4 1 2 3 4>
> myind <- paste(year, quarter, sep="q")
> myind
[1] "2000q1" "2000q2" "2000q3" "2000q4"
"2001q1" "2001q2" "2001q3"
"2001q4">
> order(myind)
[1] 1 2 3 4 5 6 7 8
(of course you must be careful with the double digits in case of months and so
on).
As all 'plm' uses is "standard" R ordering, please refer to
the general documentation of the R system for any issues.
Best wishes,
Giovanni
------------- original message -----------------
Message: 182
Date: Wed, 3 Feb 2010 10:37:13 +0100
From: "Bunny, lautloscrew.com" <bunny at lautloscrew.com>
To: r-help at r-project.org
Subject: [R] plm package index
Message-ID: <9B6A22F4-5DD0-4FCD-9A6F-AEE1E2108E36 at lautloscrew.com>
Content-Type: text/plain
Dear all,
i just wonder if there?s a way to use a two column time index field in plm
package.
the manual says the following concerning data indexing:
a character vector of length two containing the names of the individual and the
time
index,
What would y?all do with a quarterly dataset that contains one column for the
period and one for the year. Do I have to convert it to one single date field
first ?
thx in advance
matt
[[alternative HTML version deleted]]
------------- end original message ---------------
Giovanni Millo
Research Dept.,
Assicurazioni Generali SpA
Via Machiavelli 4,
34132 Trieste (Italy)
tel. +39 040 671184
fax +39 040 671160