Parker, Thomas (2010): A comparison of alternative approaches to supnorm goodness of fit tests with estimated parameters.
Preview 
PDF
MPRA_paper_22961.pdf Download (205kB)  Preview 
Abstract
Goodness of fit tests based on supnorm statistics of empirical processes have nonstandard limiting distributions when the null hypothesis is composite — that is, when parameters of the null model are estimated. Several solutions to this problem have been suggested, including the calculation of adjusted critical values for these nonstandard distributions and the transformation of the empirical process such that statistics based on the transformed process are asymptotically distributionfree. The approximation methods proposed by Durbin (1985) can be applied to compute appropriate critical values for tests based on supnorm statistics. The resulting tests have quite accurate size, a fact which has gone unrecognized in the econometrics literature. Some justification for this accuracy lies in the similar features that Durbin’s approximation methods share with the theory of extrema for Gaussian random fields and for GaussMarkov processes. These adjustment techniques are also related to the transformation methodology proposed by Khmaladze (1981) through the score function of the parametric model. Monte Carlo experiments suggest that these two testing strategies are roughly comparable to one another and more powerful than a simple bootstrap procedure.
Item Type:  MPRA Paper 

Original Title:  A comparison of alternative approaches to supnorm goodness of fit tests with estimated parameters 
Language:  English 
Keywords:  Goodness of fit test; Estimated parameters; Gaussian process; GaussMarkov process; Boundary crossing probability; Martingale transformation 
Subjects:  C  Mathematical and Quantitative Methods > C1  Econometric and Statistical Methods and Methodology: General > C14  Semiparametric and Nonparametric Methods: General C  Mathematical and Quantitative Methods > C1  Econometric and Statistical Methods and Methodology: General > C12  Hypothesis Testing: General C  Mathematical and Quantitative Methods > C4  Econometric and Statistical Methods: Special Topics > C46  Specific Distributions ; Specific Statistics 
Item ID:  22961 
Depositing User:  Thomas Parker 
Date Deposited:  30 May 2010 06:37 
Last Modified:  02 Oct 2019 11:14 
References:  S. Aki. Some test statistics based on the martingale term of the empirical distribution function. Annals of the Institute of Statistical Mathematics, 38(1):1–21, 1986. J. Bai. Testing parametric conditional distributions of dynamic models. Review of Economics and Statistics, 85(3):531–549, 2003. P. Bickel, C. Klaassen, Y. Ritov, and J. Wellner. Efficient and Adaptive Estimation for Semiparametric Models. Johns Hopkins University Press, 1993. A. Buonocore, A. Nobile, and L. Ricciardi. A new integral equation for the evaluation of firstpassagetime probability densities. Advances in Applied Probability, 19(4):784–800, 1987. A. Cabaña and E. Cabaña. Transformed empirical processes and modified KolmogorovSmirnov tests for multivariate distributions. Annals of Statistics, 25(6):2388–2409, 1997. E. del Barrio. Lectures on Empirical Processes: Theory and Statistical Applications, chapter Empirical and Quantile Processes in the Asymptotic Theory of Goodnessoffit Tests, pages 1–92. EMS Series of Lectures in Mathematics. European Mathematical Society, 2007. M. Delgado and W. Stute. Distributionfree specification tests of conditional models. Journal of Econometrics, 143(1):37–55, 2008. E. Di Nardo, A. Nobile, E. Pirozzi, and L. Ricciardi. A computational approach to firstpassagetime problems for GaussMarkov processes. Advances in Applied Probability, 33(2):453–482, 2001. J. Doob. Stochastic Processes. Wiley, 1953. J. Durbin. Boundarycrossing probabilities for the Brownian motion and Poisson processes and techniques for computing the power of the KolmogorovSmirnov test. Journal of Applied Probability, 8 (3):431–453, 1971. J. Durbin. Weak convergence of the sample distribution function when parameters are estimated. The Annals of Statistics, 1(2):279–290, 1973a. J. Durbin. Distribution Theory for Tests Based on the Sample Distribution Function. Number 9 in Regional Conference Series in Applied Mathematics. SIAM, 1973b. J. Durbin. KolmogorovSmirnov tests when parameters are estimated with applications to tests of exponentiality and tests on spacings. Biometrika, 62(1):5–22, 1975. J. Durbin. The firstpassage density of a continuous Gaussian process to a general boundary. Journal of Applied Probability, 22(1):99–122, 1985. J. Durbin, M. Knott, and C. Taylor. Components of the Cramérvon Mises statistics. II. Journal of the Royal Statistical Society, Series B (Methodological), 37(2):216–237, 1975. V. Fatalov. Asymptotics of large deviation probabilities for Gaussian fields. Journal of Contemporary Mathematical Analysis, 27(3):48–70, 1992. V. Fatalov. Asymptotics of large deviation probabilities for Gaussian fields: Applications. Journal of Contemporary Mathematical Analysis, 28(5):21–44, 1993. J. Haywood and E. Khmaladze. On distributionfree goodnessoffit testing of exponentiality. Journal of Econometrics, 143(1):5–18, 2008. Y. Hong and J. Liu. Generalized residualbased specification testing for duration models with censoring. Cornell University, 2007. Y. Hong and J. Liu. Goodnessoffit testing for duration models with censored grouped data. Cornell University, 2009. E. Khmaladze. The use of ω2 tests for testing parametric hypotheses. Theory of Probability and its Applications, 24(2):283–301, 1979. E. Khmaladze. Martingale approach in the theory of goodnessoffit tests. Theory of Probability and its Applications, 26(2):240–257, 1981. E. Khmaladze and H. Koul. Martingale transforms goodnessoffit tests in regression models. The Annals of Statistics, 32(3):995–1034, 2004. E. Khmaladze and H. Koul. Goodnessoffit problem for errors in nonparametric regression: Distribution free approach. The Annals of Statistics, 37(6A):3165–3185, 2009. R. Koenker and Z. Xiao. Inference on the quantile regression process. Econometrica, 70(4):1583–1612, 2002. H. Koul. Weighted Empirical Processes in Dynamic Nonlinear Models, volume 166 of Lecture Notes in Statistics. Springer, 2nd edition, 2002. H. Koul. Model diagnostics via martingale transforms: A brief review. In J. Fan and H. Koul, editors, Frontiers in Statistics, chapter 9, pages 183–206. Imperial College Press, 2006. H. Koul and L. Sakhanenko. Goodnessoffit testing in regression: A finite sample comparison of bootstrap methodology and Khmaladze transformation. Statistics & Probability Letters, 74(3):290–302, 2005. E. Kulinskaya. Coefficients of the asymptotic distribution of the KolmogorovSmirnov statistic when parameters are estimated. Journal of Nonparametric Statistics, 5(1):43–60, 1995. B. Li. Asymptotically distributionfree goodnessoffit testing: A unifying view. Econometric Reviews, 28 (6):632–657, 2009. R. Loynes. The empirical distribution function of residuals from generalised regression. The Annals of Statistics, 8(2):285–298, 1980. G. Martynov. Goodnessoffit tests for the Weibull and Pareto distributions. Paper presented at the Sixth International Conference on Mathematical Methods in Reliability, 2009. M. Matsui and A. Takemura. Empirical characteristic function approach to goodnessoffit tests for the Cauchy distribution with parameters estimated by MLE or EISE. Annals of the Institute of Statistical Mathematics, 57(1):183–199, 2005. C. Mehr and J. McFadden. Certain properties of Gaussian processes and their firstpassage times. Journal of the Royal Statistical Society, Series B (Methodological), 27(3):505–522, 1965. G. Neuhaus. Weak Convergence Under Contiguous Alternatives when Parameters are Estimated: the Dk approach, volume 566 of Lecture Notes in Mathematics, pages 68–82. Springer, 1976. G. Peskir. On integral equations arising in the firstpassage problem for Brownian motion. Journal of Integral Equations and Applications, 14(4):397–423, 2002. V. Piterbarg. Asymptotic Methods in the Theory of Gaussian Processes and Fields, volume 148 of Translations of Mathematical Monographs. American Mathematical Society, 1996. W. Press, S. Teukolsky, W. Vetterling, and B. Flannery. Numerical Recipes in Fortran 77: The Art of Scientific Computing. Cambridge University Press, 2nd edition, 2001. D. Rabinovitz. Estimating Durbin’s approximation. Biometrika, 80(3):671–680, 1993. J. Romano. A bootstrap revival of some nonparametric distance tests. Journal of the American Statistical Association, 83(403):698–708, 1988. G. Shorack and J. Wellner. Empirical Processes with Applications to Statistics. Wiley, 1986. K. Song. Testing semiparametric conditional moment restrictions using conditional martingale transforms. Journal of Econometrics, 154(1):74–84, 2010. Y. Tyurin. On the limit distribution of KolmogorovSmirnov statistics for a composite hypothesis. Mathematics of the USSR — Izvestiya, 25(3):619–646, 1985. A. van der Vaart and J. Wellner. Weak Convergence and Empirical Processes. Springer, 1996. 
URI:  https://mpra.ub.unimuenchen.de/id/eprint/22961 
Available Versions of this Item

A comparison of alternative approaches to supnorm goodness of git gests with estimated parameters. (deposited 28 May 2010 09:36)
 A comparison of alternative approaches to supnorm goodness of fit tests with estimated parameters. (deposited 30 May 2010 06:37) [Currently Displayed]