Dear Danice,
as far as I know, three-way panels are not considered in the
econometrics literature (two dimensions make things complicated enough
already). They are also not implemented in plm.
You might find support for more elaborate nesting structures in the nlme
and lme4 packages. Yet, as the empirical question is not clear from what
you say, you might as well want to use separate regressions,
destination/origin/time dummies (possibly interacted with coefficients)
and so on.
Best wishes,
Giovanni
------- original message ------------
Message: 111
Date: Tue, 13 Jul 2010 07:14:06 +0100
From: danice ng <danicenyl at gmail.com>
To: r-help at r-project.org
Subject: [R] Three-way Panel Data Analysis
Message-ID:
<AANLkTimE7a8OYDgOtqR-Qw0MAUw8iqN7RVkNc9Csxz7B at mail.gmail.com>
Content-Type: text/plain
Dear R users,
I have panel data on the amount of money spent by travellers from 8
origin
countries in 4 destinations. I would like to carry out analysis for
destinations, origins and time. However, it seems to me that the package
"plm" can only esitmate two-way panel data (indexed by a
two-dimensional
array). Any suggestions would be greatly appreciated.
Thank you.
Best regards,
Danice
[[alternative HTML version deleted]]
------------------------------
Giovanni Millo
Research Dept.,
Assicurazioni Generali SpA
Via Machiavelli 4,
34132 Trieste (Italy)
tel. +39 040 671184
fax +39 040 671160