(1) Departamento de Estadística, Investigación Operativa y Cálculo Numérico Facultad de Ciencias, U.N.E.D. Paseo Senda del Rey 9, 28040 Madrid, Spain
Abstract:
Many test statistics follow a χ2 distribution because a normal model is assumed as underlying distribution. In this paper we obtain good analytic approximations for the p-value and the critical value of χ2 tests when the underlying distribution is close but different from the normal model. With these approximations we study the robustness of validity of χ2 tests