Several new tests are proposed for examining the adequacy of a family of pa
rametric models against large nonparametric alternatives. These tests forma
lly check if the bias vector of residuals from parametric fts is negligible
by using the adaptive Neyman test and other methods. The testing procedure
s formalize the traditional model diagnostic tools based on residual plots.
We examine the rates of contiguous alternatives that can be detected consi
stently by the adaptive Neyman test. Applications of the procedures to the
partially linear models are thoroughly discussed. Our simulation studies sh
ow that the new testing procedures are indeed powerful and omnibus. The pow
er of the proposed tests is comparable to the F-test statistic even in the
situations where the F test is known to be suitable and can be far more pow
erful than the F-test statistic in other situations. An application to test
ing linear models versus additive models is also discussed.