<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body>
<div>
<meta content="text/html; charset=us-ascii">
</div>
<div dir="auto" style="text-align:left">Dear all,</div>
<div dir="auto" style="text-align:left"><br>
</div>
<div dir="auto" style="text-align:left">Currently, I am trying to establish a statistical relationship between the occurrence of a land cover type and a set of local and global climate variables.</div>
<div dir="auto" style="text-align:left"><br>
</div>
<div dir="auto" style="text-align:left">So far, I tried amongst others machine learning as well as bayesian and other modelling techniques. Unfortunately, those approaches proved so computationally demanding that I was only able to use a subset of my datapoints
that does not really cover the variation.</div>
<div dir="auto" style="text-align:left"><br>
</div>
<div dir="auto" style="text-align:left">In r.regression.multi I can use the entire map which is great. However, I am not sure if it is statistically sound to use r.regression.multi for logistic regression (0/1 data). Is the module appropriate for that purpose?</div>
<div dir="auto" style="text-align:left"><br>
</div>
<div dir="auto" style="text-align:left">If not, could it be extended? </div>
<div dir="auto" style="text-align:left"><br>
</div>
<div dir="auto" style="text-align:left">Cheers</div>
<div dir="auto" style="text-align:left">Stefan</div>
<div dir="auto" style="text-align:left"><br>
</div>
<div dir="auto" style="text-align:left"><br>
</div>
<div><br>
</div>
<div><br>
</div>
</body>
</html>