s-news
[Top] [All Lists]

Re: Moran's I for residuals

To: kate.bassil@utoronto.ca
Subject: Re: Moran's I for residuals
From: Volker Bahn <lochapoka@web.de>
Date: Fri, 04 Apr 2008 11:14:42 -0400
Cc: s-news@lists.biostat.wustl.edu
In-reply-to: <20080402105724.pytpf7o8swocskgw@webmail.utoronto.ca>
References: <20080402105724.pytpf7o8swocskgw@webmail.utoronto.ca>
User-agent: Thunderbird 2.0.0.12 (Windows/20080213)
Hi Kate,

You can find extensive code for doing this in the ecological archives supplemental files for Jeremy W. Lichstein, Theodore R. Simons, Susan A. Shriner, and Kathleen E. Franzreb. 2002. Spatial autocorrelation and autoregressive models in ecology. Ecological Monographs 72:445-463.

at http://www.esapubs.org/archive/mono/M072/007/default.htm

The short answer is that you need to do the randomization for finding the reference distribution of Morin's I on the original data and then run the model on each randomization and calculate Morin's I for each set of the resulting residuals rather than simply running randomizations of the residuals of the model on the original data.

HTH

Volker


kate.bassil@utoronto.ca wrote:
<div class="moz-text-flowed" style="font-family: -moz-fixed">Can anyone tell me how to calculate Moran's I for regression residuals using s-plus? I realize I can calculate it for the raw data using 'spatial.cor' but thought a modification was necessary given I would like to apply it to regression residuals.
Thank you.

Kate Bassil
University of Toronto

</div>



<Prev in Thread] Current Thread [Next in Thread>