Testing the Significance of Categorical Variables in 
Nonparametric Regression Models 


Abstract


In this paper we propose a test for the significance of categorical
variables in nonparametric regression models. The test is fully-data
driven and employs cross-validated smoothing parameter selection while
the null distribution of the test is obtained via pivotal
bootstrapping methods. Simulations reveal that the test has correct
size and has power that increases with both the departure from the
null and the sample size. An application is considered.

(This is joint work with Q. Li).