Summarize the result of a multiple comparison statistical test in a table


tabularTestSummary(ph, columns = c("pvalue"))


The input testMultiple from which the table is generated
A vector indicating the metrics that will be shown in the table


an extabular object


This function builds a table from a testMultiple object, either control or pairwise. The htpotheses are added and compared in the table showing the methods and a range of different metrics than can be added to the table. Also the table shows information about rejected hypotheses.


# First we create an experiment from the wekaExperiment problem and prepare # it to apply the test: experiment <- expCreate(wekaExperiment, name="test", parameter="fold") experiment <- expReduce(experiment, "fold", mean) experiment <- expInstantiate(experiment, removeUnary=TRUE) # Then we perform a a testMultiplePairwise test procedure test <- testMultipleControl(experiment, "accuracy", "min") # Different tables can be obtained by using a range of metrics tabularTestSummary(test, c("pvalue"))
$testMultiple method pvalue 7 OneR,yes NA 3 OneR,no 7.093881e-01 6 NaiveBayes,yes 1.596898e-01 2 NaiveBayes,no 8.367640e-02 5 J48,yes 1.295346e-02 8 RandomForest,yes 8.359247e-04 1 J48,no 7.791896e-05 4 RandomForest,no 1.551336e-06
tabularTestSummary(test, c("rank", "pvalue", "wtl"))
$testMultiple method rank pvalue win tie loss OneR,yes OneR,yes 2.200000 NA NA NA NA OneR,no OneR,no 2.533333 7.093881e-01 3 12 0 NaiveBayes,yes NaiveBayes,yes 3.766667 1.596898e-01 10 0 5 NaiveBayes,no NaiveBayes,no 4.166667 8.367640e-02 10 0 5 J48,yes J48,yes 4.833333 1.295346e-02 14 0 1 RandomForest,yes RandomForest,yes 5.566667 8.359247e-04 14 0 1 J48,no J48,no 6.100000 7.791896e-05 15 0 0 RandomForest,no RandomForest,no 6.833333 1.551336e-06 15 0 0

Back to top

Built by staticdocs. Styled with bootstrap.