This paper develops a test of autocorrelation in the presence of heter
oskedasticity of unknown form in the nonlinear regression model, The t
est statistic is based on the sample autocovariance of the residuals s
tandardized by a nonparametric kernel estimate of the unknown heterosk
edasticity function, Under appropriate conditions, the test statistic
is shown to have a limiting chi-square distribution, Local power and c
onsistency results for the test are also established, Monte Carlo expe
riments show that the test has reasonable size performance and general
ly dominates some of the existing tests in terms of finite-sample powe
r.