- statsmodels.stats.diagnostic.linear_harvey_collier(res, order_by=None, skip=None)¶
Harvey Collier test for linearity
The Null hypothesis is that the regression is correctly modeled as linear.
A results instance from a linear regression.
Integer array specifying the order of the residuals. If not provided, the order of the residuals is not changed. If provided, must have the same number of observations as the endogenous variable.
The number of observations to use for initial OLS, if None then skip is set equal to the number of regressors (columns in exog).
Recursive OLS residual calculation used in the test.
This test is a t-test that the mean of the recursive ols residuals is zero. Calculating the recursive residuals might take some time for large samples.