An equality test across nonparametric regressions

We propose a procedure for testing the equality of several nonparametric mu ltivariate regressions. We allow the regressors' designs and the number of observations to differ across subsamples. The division into subsamples is d efined through a variable C which can be either fixed or random. For a rand om C, our procedure is a general test or significance for qualitative varia bles in a nonparametric regression. For a fixed C, our procedure provides a 'nonparametric analysis of covariance', which is valid for cross-section o r panel data. In both cases, the test is a one-sided normal test acid is co nsistent against all alternatives. (C) 2001 Elsevier Science S.A. All right s reserved.