3.8.6.3.6. statsmodels.nonparametric.kernel_regression.TestRegCoefC¶
-
class
statsmodels.nonparametric.kernel_regression.TestRegCoefC(model, test_vars, nboot=400, nested_res=400, pivot=False)[source]¶ Significance test for continuous variables in a nonparametric regression.
The null hypothesis is
dE(Y|X)/dX_not_i = 0, the alternative hypothesis isdE(Y|X)/dX_not_i != 0.Parameters: model: KernelReg instance
This is the nonparametric regression model whose elements are tested for significance.
test_vars: tuple, list of integers, array_like
index of position of the continuous variables to be tested for significance. E.g. (1,3,5) jointly tests variables at position 1,3 and 5 for significance.
nboot: int
Number of bootstrap samples used to determine the distribution of the test statistic in a finite sample. Default is 400
nested_res: int
Number of nested resamples used to calculate lambda. Must enable the pivot option
pivot: bool
Pivot the test statistic by dividing by its standard error Significantly increases computational time. But pivot statistics have more desirable properties (See references)
Notes
This class allows testing of joint hypothesis as long as all variables are continuous.
References
Racine, J.: “Consistent Significance Testing for Nonparametric Regression” Journal of Business & Economics Statistics.
Chapter 12 in [1].
Attributes
sig: str The significance level of the variable(s) tested “Not Significant”: Not significant at the 90% confidence level Fails to reject the null “*”: Significant at the 90% confidence level “**”: Significant at the 95% confidence level “***”: Significant at the 99% confidence level