[GRASS-SVN] r73784 - grass-addons/grass7/raster/r.mregression.series

[svn_grass at osgeo.org]

Author: DmitryKolesov
Date: 2018-12-10 08:16:36 -0800 (Mon, 10 Dec 2018)
New Revision: 73784

Modified:
   grass-addons/grass7/raster/r.mregression.series/r.mregression.series.html
Log:
Fix html desctiption in r.mregression.series

Modified: grass-addons/grass7/raster/r.mregression.series/r.mregression.series.html
===================================================================
--- grass-addons/grass7/raster/r.mregression.series/r.mregression.series.html	2018-12-10 15:45:05 UTC (rev 73783)
+++ grass-addons/grass7/raster/r.mregression.series/r.mregression.series.html	2018-12-10 16:16:36 UTC (rev 73784)
@@ -2,7 +2,7 @@
 
 <em>r.mregression.series</em> is a module to calculate multiple 
 linear regression parameters between several time series, e.g. NDVI and
-elevation, precipitation. It uses the <em>python-statmodels</em> package.
+elevation, precipitation. It s a front-end for models from  <em>python-statmodels</em> package.
 <p>
 The module makes each output cell value a function of the values
 assigned to the corresponding cells in the input raster map series.
@@ -11,18 +11,23 @@
 
 The module assumes a simple linear regression of the form
 <div class="code"><pre>
-    Y = b1 * X1 + b2 * X2 + ... + bn * Xn 
+    Y(t) = b1 * X1(t) + b2 * X2(t) + ... + bn * Xn(t) 
 </pre></div>
 <p>
 
+The module uses two models: ordinary least squares and robust linear models.
+
 <h2>NOTES</h2>
+The module performs multiple linear regression, use 
+<a href="r.regression.series.html">r.regression.series</a> for regression with one predictor.
 
-The number of predictor variables (<em>X</em> maps) needs to be the same in 
-each (time) series (see examples below). If the different time series have irregular 
-time intervals, NULL raster maps can be inserted into time series to
-make time intervals equal. 
 <p>
-The list of inputs for each cell (including NULLs) is passed to the 
+The number of predictor variables (<em>X</em> maps) must be the same in 
+each (time) series (see examples below). If the different predictors have 
+different or irregular time intervals, NULL raster maps can be inserted 
+into time series to make time intervals equal. 
+<p>
+The list of raster inputs (including NULLs) is passed to the 
 regression function. The function computes the parameters over the 
 non-NULL values, producing a NULL result only if there aren't enough 
 non-NULL values for computing.
@@ -30,9 +35,9 @@
 
 <h2>EXAMPLES</h2>
 The most important paramether is <em>samples</em>; it provides the list
-of <em>X</em> and <em>Y</em> maps. The parameter is the name of csv file
+of <em>Y</em> and <em>X</em> maps. The parameter is the name of csv file
 of the next structure: the first line is a header, other lines provide names
-of the <em>X</em> and <em>Y</em> maps. The header contains the names of the
+of the <em>Y</em> and <em>X</em> maps. The header contains the names of the
 input and output variables.
 <p>
 For example the csv file for regression between NDVI and (elevation, 
@@ -57,6 +62,7 @@
 the names of the regression coefficients will be "coef.elevation"
 and "coef.precipitation".
 
+
 <div class="code"><pre>
 r.mregression.series samples=settings result_prefix="coef."
 </pre></div>
@@ -84,7 +90,7 @@
 produces three raster maps: "coef.offset", "coef.elevation", "coef.precipitation".    
 
 <h3>EXAMPLE 1</h3>
-We create test data for the example. Suppose we have five <em>Y</em> maps and
+Create test data for the example. Suppose we have five <em>Y</em> maps and
 5 pairs of predictor <em>X</em> = <em>(x1, x2)</em> maps.
 <p>
 Create <em>X</em> variables (random numbers):