A Mean Likelihood Ratio Speci cation Test
نویسنده
چکیده
This paper considers the problem of speci cation testing in general parametric models and shows that for a wide class of models the hypothesis of correct speci cation is equivalent to a continuum of moment equalities. Using these moment equalities we construct a class of speci cation tests that have correct asymptotic size in general parametric models, including stationary time series models, and that are consistent when the above equivalence holds. We show that the proposed tests have power against √ T -local alternatives and compare them to previously proposed consistent tests of distributional speci cation, both from a theoretical perspective and in simulation. JEL Classi cation: C52
منابع مشابه
Maximum Likelihood Estimation and Uniform Inference with Sporadic Identication Failure
This paper analyzes the properties of a class of estimators, tests, and con dence sets (CSs) when the parameters are not identi ed in parts of the parameter space. Speci cally, we consider estimator criterion functions that are sample averages and are smooth functions of a parameter : This includes log likelihood, quasi-log likelihood, and least squares criterion functions. We determine the as...
متن کاملMaximum likelihood estimation and uniform inference with sporadic identification failure
This paper analyzes the properties of a class of estimators, tests, and con dence sets (CSs) when the parameters are not identi ed in parts of the parameter space. Speci cally, we consider estimator criterion functions that are sample averages and are smooth functions of a parameter : This includes log likelihood, quasi-log likelihood, and least squares criterion functions. We determine the as...
متن کاملLM Test of Neglected Correlated Random E¤ects and Its Application
Econometric model speci cation is usually tested by exploiting one of the three principles, including (i) the Wald test, which is based on the asymptotic distribution of parameter estimates; (ii) the Likelihood Ratio (LR) test; and (iii) the Lagrange Multiplier (LM) procedure, which is based on the derivative of the log likelihood (score) imposing the hypothesis.1 The LM procedure seems to domi...
متن کاملSpatial Panels: Random Components vs. Fixed E¤ects
This paper investigates spatial panel data models with a space-time lter in disturbances. We consider their estimation by both xed e¤ects and random e¤ects speci cations. With a between equation properly de ned, the di¤erence of the random vs. xed e¤ects models can be highlighted. We show that the random e¤ects estimate is a pooling of the within and between estimates. A Hausman type speci c...
متن کاملAutomating Formal Speciication-based Testing to Appear: Tapsoft '97
This paper presents a technique for automatically generating logical schemata that specify groups of black-box test cases from formal speci cations containing universal and existential quanti cation. These schemata are called test frames. Previous automated techniques have dealt with languages based on propositional logic. Since this new technique deals with quanti cation it can be applied to m...
متن کامل