[R-sig-ME] Which package handles both auto-correlation and, heteroskedacity?

[Highland Statistics Ltd]

------------------------------

Message: 4
Date: Mon, 14 May 2012 10:00:20 -0500
From: Luna<lunamoonmoon at gmail.com>
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] Which package handles both auto-correlation and
	heteroskedacity?
Message-ID:
	<CAPNjSFbxxYYnJcPgE4bgFEXzKhbhvRSQZe0SMM5o6Uf+E8LtOQ at mail.gmail.com>
Content-Type: text/plain

Which package handles both auto-correlation and heteroskedacity?

Hi all,


Using the terminology on this webpage:

http://www.nyu.edu/its/pubs/connect/fall03/yaffee_primer.html

We are in need of doing the "Dynamic Panel Models".

We would like to incorporate "auto-correlation" along the time axis and the
"heteroskedacity" cross-sectionally.

> From literature reading, the "time series" concept is sometimes called
"longitudinal".

Could you please recommend a package for that (nlme or lme4 or others)?

And could you please point us to a concrete example on this?

Thanks a lot!







You already answered your own question....nlme. Chapter 5 of Pinheiro and Bates contains plenty of examples.
So does our 2009 volume (Zuur et al 2009). Or a recent paper in MEE:

Neglected biological patterns in the residuals. A behavioural ecologist’s guide to co-operating with heteroscedasticity
Ian R. Cleasby&  Shinichi Nakagawa

Alain








------------------------------

_______________________________________________
R-sig-mixed-models mailing list
R-sig-mixed-models at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models


End of R-sig-mixed-models Digest, Vol 65, Issue 35