5.9. statsmodels.sandbox.multilinear¶
Analyze a set of multiple variables with a linear models
- multiOLS:
- take a model and test it on a series of variables defined over a pandas dataset, returning a summary for each variable
- multigroup:
- take a boolean vector and the definition of several groups of variables and test if the group has a fraction of true values higher than the rest. It allows to test if the variables in the group are significantly more significant than outside the group.
5.9.1. Functions¶
iteritems(obj, **kwargs) |
replacement for six’s iteritems for Python2/3 compat |
multiOLS(model, dataframe[, column_list, ...]) |
apply a linear model to several endogenous variables on a dataframe |
multigroup(pvals, groups[, exact, keep_all, ...]) |
Test if the given groups are different from the total partition. |