|
80
|
1 <tool id="partialRsq" name="Compute partial R square" version="1.1.0">
|
|
|
2 <description> </description>
|
|
|
3 <expand macro="requirements" />
|
|
|
4 <macros>
|
|
|
5 <import>statistic_tools_macros.xml</import>
|
|
|
6 </macros>
|
|
|
7 <command interpreter="python">
|
|
|
8 partialR_square.py
|
|
|
9 $input1
|
|
|
10 $response_col
|
|
|
11 $predictor_cols
|
|
|
12 $out_file1
|
|
|
13 1>/dev/null
|
|
|
14 </command>
|
|
|
15 <inputs>
|
|
|
16 <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" />
|
|
|
18 <param name="predictor_cols" label="Predictor columns (X)" type="data_column" data_ref="input1" multiple="true">
|
|
|
19 <validator type="no_options" message="Please select at least one column."/>
|
|
|
20 </param>
|
|
|
21 </inputs>
|
|
|
22 <outputs>
|
|
|
23 <data format="input" name="out_file1" metadata_source="input1" />
|
|
|
24 </outputs>
|
|
|
25 <tests>
|
|
|
26 <!-- Test data with vlid values -->
|
|
|
27 <test>
|
|
|
28 <param name="input1" value="regr_inp.tabular"/>
|
|
|
29 <param name="response_col" value="3"/>
|
|
|
30 <param name="predictor_cols" value="1,2"/>
|
|
|
31 <output name="out_file1" file="partialR_result.tabular"/>
|
|
|
32 </test>
|
|
|
33
|
|
|
34 </tests>
|
|
|
35 <help>
|
|
|
36
|
|
|
37 .. class:: infomark
|
|
|
38
|
|
|
39 **TIP:** If your data is not TAB delimited, use *Edit Datasets->Convert characters*
|
|
|
40
|
|
|
41 -----
|
|
|
42
|
|
|
43 .. class:: infomark
|
|
|
44
|
|
|
45 **What it does**
|
|
|
46
|
|
|
47 This tool computes the Partial R squared for all possible variable subsets using the following formula:
|
|
|
48
|
|
|
49 **Partial R squared = [SSE(without i: 1,2,...,p-1) - SSE (full: 1,2,..,i..,p-1) / SSE(without i: 1,2,...,p-1)]**, which denotes the case where the 'i'th predictor is dropped.
|
|
|
50
|
|
|
51
|
|
|
52
|
|
|
53 In general, **Partial R squared = [SSE(without i: 1,2,...,p-1) - SSE (full: 1,2,..,i..,p-1) / SSE(without i: 1,2,...,p-1)]**, where,
|
|
|
54
|
|
|
55 - SSE (full: 1,2,..,i..,p-1) = Sum of Squares left out by the full set of predictors SSE(X1, X2 … Xp)
|
|
|
56 - SSE (full: 1,2,..,i..,p-1) = Sum of Squares left out by the set of predictors excluding; for example, if we omit the first predictor, it will be SSE(X2 … Xp).
|
|
|
57
|
|
|
58
|
|
|
59 The 4 columns in the output are described below:
|
|
|
60
|
|
|
61 - Column 1 (Model): denotes the variables present in the model
|
|
|
62 - Column 2 (R-sq): denotes the R-squared value corresponding to the model in Column 1
|
|
|
63 - Column 3 (Partial R squared_Terms): denotes the variable/s for which Partial R squared is computed. These are the variables that are absent in the reduced model in Column 1. A '-' in this column indicates that the model in Column 1 is the Full model.
|
|
|
64 - Column 4 (Partial R squared): denotes the Partial R squared value corresponding to the variable/s in Column 3. A '-' in this column indicates that the model in Column 1 is the Full model.
|
|
|
65
|
|
|
66 *R Development Core Team (2010). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org.*
|
|
|
67
|
|
|
68 </help>
|
|
|
69 </tool>
|
