Abstract

This article considers two conditional moment tests for neglected nonlinearity in regression models and examines their finite sample performance. The two tests are the nonparametric kernel test by Li and Wang (1998) and Zheng (1996) and the neural network test of White (1989). The article examines an asymptotic test, a naive bootstrap test, and a wild bootstrap test for weakly dependent time series and independent data.

Recommended Citation

Tae-Hwy Lee (2001) "Neural Network Test and Nonparametric Kernel Test for Neglected Nonlinearity in Regression Models ", Studies in Nonlinear Dynamics & Econometrics: Vol. 4: No. 4, Article 1.
http://www.bepress.com/snde/vol4/iss4/art1

Related Files

lee_code.zip (6 kB)
Code