Biometrika Advance Access originally published online on November 19, 2007
Biometrika 2007 94(4):992-998; doi:10.1093/biomet/asm065
Miscellanea |
Use of the Gibbs Sampler to Obtain Conditional Tests, with Applications
Department of Statistics and Actuarial Science, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada lockhart{at}stat.sfu.ca
Instituto de Investigaciones en Matemáticas Aplicadas y en Sistemas, Universidad Nacional Autónoma de México, Mexico federico{at}sigma.iimas.unam.mx
Department of Statistics and Actuarial Science, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada stephens{at}stat.sfu.ca
Received for publication 1 March 2006. Revision received 1 March 2007.
A random sample is drawn from a distribution which admits a minimal sufficient statistic for the parameters. The Gibbs sampler is proposed to generate samples, called conditionally sufficient or co-sufficient samples, from the conditional distribution of the sample given its value of the sufficient statistic. The procedure is illustrated for the gamma distribution. Co-sufficient samples may be used to give exact tests of fit; for the gamma distribution these are compared for size and power with approximate tests based on the parametric bootstrap.
Key Words: Empirical distribution function test Goodness-of-fit test Parametric bootstrap Rao–Blackwell Sufficient statistic
References
-
Cheng R. C. H. Generation of inverse Gaussian variates with given sample mean and dispersion. Appl. Statist. (1984) 33:309–16.[CrossRef]
Engen S., Lillegård M. Stochastic simulations conditioned on sufficient statistics. Biometrika (1997) 84:235–40.
Gelfand A. E., Smith A. F. M. Sampling-based approaches to calculating marginal densities. J. Am. Statist. Assoc. (1990) 85:398–409.[CrossRef][Web of Science]
González-Barrios J., O'Reilly F., Rueda R. Goodness-of-fit for discrete random variables using the conditional density. Metrika (2006) 64:77–94.[CrossRef][Web of Science]
Gürtler N., Henze N. Recent and classical goodness-of-fit tests for the Poisson distribution. J. Statist. Plan. Infer. (2000) 90:207–25.[CrossRef]
Lindqvist B. H., Taraldsen G. Monte Carlo conditioning on a sufficient statistic. Biometrika (2005) 92:451–64.
Lindqvist B. H., Taraldsen G., Lillegård M., Engen S. A counterexample to a claim about stochastic simulations. Biometrika (2003) 90:489–90.
O'Reilly F., Gracia-Medrano L. On the conditional distribution of goodness-of-fit tests. Commun. Statist. A (2006) 35:541–9.
Stephens M. A. EDF statistics for goodness of fit and some comparisons. J. Am. Statist. Assoc. (1974) 69:730–7.[CrossRef][Web of Science]
Stephens M. A. Tests based on EDF statistics. In: Goodness of Fit Techniques—D'Agostino R. B., Stephens M. A, eds. (1986) New York: Marcel Dekker. 97–193.
Tierney L. Markov chains for exploring posterior distributions (with Discussion). Ann. Statist. (1994) 22:1701–62.[CrossRef]
| ||||||||||||||||||||||||||||||||||||||||||||||||||