In the document for function seqecmpgroup(): "subsequences are then ordered by decreasing discriminant power."
Just wondering what is the method to compute the value? It looks like the 'statistic' column.
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
The test is performed by means of a Pearson chi-square independence test on the table obtained by cross tabulating the presence/absence of the subsequence with the group variable. The values returned by seqecmpgroup as statistic are the Pearson independence chi-squares.
The smaller the p-value of the test, the more powerful is the subsequence for discriminating between the groups.
The p-value is calculated with a Bonferroni correction when method = "bonferroni" and without correction when method = "chisq" (default value).
Note that the same Bonferroni correction is applied for each subsequence. Therefore, applying or not the correction does not change the ranking of the subsequences according to their discriminating power.
