<p>Excellent! Can't wait to test it out. Already have several applications in mind.</p>
<div class="gmail_quote">On Nov 8, 2011 1:28 AM, "Markus Metz" <<a href="mailto:markus.metz.giswork@googlemail.com">markus.metz.giswork@googlemail.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Dylan Beaudette wrote:<br>
> On Friday, October 07, 2011, Markus Metz wrote:<br>
>> There is a new module r.regression.multi as grass7 add-on to calculate<br>
>> multiple regressions with raster maps. The motivation for this module<br>
>> is to calculate regression coefficients and statistics for very large<br>
>> datasets, too large for e.g. R. The module uses less than 3 MB memory<br>
>> for 400 million cells with one response variable and 8 predictors.<br>
>> Including residuals, this makes a total of 4 billion numbers.<br>
>> Calculation takes about 4 minutes for this dataset on my laptop. In<br>
>> addition to the slope estimates, statistics provided are R squared,<br>
>> adjusted R squared, F, AIC, corrected AIC, BIC for the full model, and<br>
>> F, AIC, corrected AIC, BIC for each predictor. Results are identical<br>
>> to those produced by R (with smaller test datasets).<br>
>><br>
>> Markus M<br>
>> _______________________________________________<br>
>> grass-stats mailing list<br>
>> <a href="mailto:grass-stats@lists.osgeo.org">grass-stats@lists.osgeo.org</a><br>
>> <a href="http://lists.osgeo.org/mailman/listinfo/grass-stats" target="_blank">http://lists.osgeo.org/mailman/listinfo/grass-stats</a><br>
>><br>
><br>
> Very cool. Can this module fit models from point data, and produce predictions<br>
> in the form of a raster?<br>
<br>
Yes, as of r49130. Residuals are also available as output raster.<br>
<br>
Markus M<br>
</blockquote></div>