Mercurial > repos > iuc > rpy_statistics_collection

comparison logistic_regression_vif.xml @ 1:2e7bc1bb2dbe draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author iuc
date Fri, 09 Jan 2015 12:56:07 -0500
parents ffcdde989859
children
comparison
equal deleted inserted replaced
0:ffcdde989859 1:2e7bc1bb2dbe
3 <expand macro="requirements" /> 3 <expand macro="requirements" />
4 <macros> 4 <macros>
5 <import>statistic_tools_macros.xml</import> 5 <import>statistic_tools_macros.xml</import>
6 </macros> 6 </macros>
7 <command interpreter="python"> 7 <command interpreter="python">
8 logistic_regression_vif.py 8 <![CDATA[
9 logistic_regression_vif.py
9 $input1 10 $input1
10 $response_col 11 $response_col
11 $predictor_cols 12 $predictor_cols
12 $out_file1 13 $out_file1
13 1>/dev/null 14 1>/dev/null
15 ]]>
14 </command> 16 </command>
15 <inputs> 17 <inputs>
16 <param format="tabular" name="input1" type="data" label="Select data" help="Dataset missing? See TIP below."/> 18 <param format="tabular" name="input1" type="data" label="Select data" help="Dataset missing? See TIP below."/>
17 <param name="response_col" label="Response column (Y)" type="data_column" data_ref="input1" numerical="True"/> 19 <param name="response_col" label="Response column (Y)" type="data_column" data_ref="input1" numerical="True"/>
18 <param name="predictor_cols" label="Predictor columns (X)" type="data_column" data_ref="input1" numerical="True" multiple="true" > 20 <param name="predictor_cols" label="Predictor columns (X)" type="data_column" data_ref="input1" numerical="True" multiple="true" >
31 <output name="out_file1" file="logreg_out2.tabular"/> 33 <output name="out_file1" file="logreg_out2.tabular"/>
32 34
33 </test> 35 </test>
34 </tests> 36 </tests>
35 <help> 37 <help>
38 <![CDATA[
36 39
37 40
38 .. class:: infomark 41 .. class:: infomark
39 42
40 **TIP:** If your data is not TAB delimited, use *Edit Datasets-&gt;Convert characters* 43 **TIP:** If your data is not TAB delimited, use *Edit Datasets->Convert characters*
41 44
42 ----- 45 -----
43 46
44 .. class:: infomark 47 .. class:: infomark
45 48
65 - Pseudo R-squared: the proportion of model improvement from null model 68 - Pseudo R-squared: the proportion of model improvement from null model
66 - p-value: p-value for the z-test of the null hypothesis that the corresponding slope is equal to zero against the two-sided alternative. 69 - p-value: p-value for the z-test of the null hypothesis that the corresponding slope is equal to zero against the two-sided alternative.
67 - Coefficient indicates log ratio of (probability to be class 1 / probability to be class 0) 70 - Coefficient indicates log ratio of (probability to be class 1 / probability to be class 0)
68 71
69 - This tool also provides **Variance Inflation Factor or VIF** which quantifies the level of multicollinearity. The tool will automatic generate VIF if the model has more than one predictor. The higher the VIF, the higher is the multicollinearity. Multicollinearity will inflate standard error and reduce level of significance of the predictor. In the worst case, it can reverse direction of slope for highly correlated predictors if one of them is significant. A general thumb-rule is to use those predictors having VIF lower than 10 or 5. 72 - This tool also provides **Variance Inflation Factor or VIF** which quantifies the level of multicollinearity. The tool will automatic generate VIF if the model has more than one predictor. The higher the VIF, the higher is the multicollinearity. Multicollinearity will inflate standard error and reduce level of significance of the predictor. In the worst case, it can reverse direction of slope for highly correlated predictors if one of them is significant. A general thumb-rule is to use those predictors having VIF lower than 10 or 5.
70 - **vif** is calculated by 73 - **vif** is calculated by
71 - First, regressing each predictor over all other predictors, and recording R-squared for each regression. 74 - First, regressing each predictor over all other predictors, and recording R-squared for each regression.
72 - Second, computing vif as 1/(1- R_squared) 75 - Second, computing vif as 1/(1- R_squared)
73 76
77 ]]>
74 </help> 78 </help>
75 </tool> 79 </tool>