... hypothe-sis is rejected much more often than the nominal size of the test). See, for example,Davidson and MacKinnon ( 199 3), Bera and McKenzie ( 198 6), Orme ( 199 0), and Chesher and Spady ( 199 1).The ... Gallant ( 198 7), Gallant and White ( 198 8), and White ( 199 4). See Andrews ( 198 9) for innovative suggestionsfor using local power analysis in applied work.12.7 Optimization MethodsIn this section ... calculation of standard errors, confidence intervals, and p-values for test statistics, and we canget a good idea of the amount of finite-sample bias in the estimation method. In ad-dition, under...