Research ArticleOpen AccessGoogle Scholar indexed
The Permutation Test as an Ancillary Procedure for Comparing Zero-Inflated Continuous Distributions
CSIRO Departments of Plant Science and Mathematics & Statistics, South Dakota State University, Brookings, SD 57007
Office of Health Data and Research, Mississippi State Department of Health,570 East Woodrow Wilson, Jackson, MS 39215-1700
Pennington Biomedical Research Center, Louisiana State University System, 6400 Perkins Road,Baton Rouge, LA 70808
- 1 CSIRO Departments of Plant Science and Mathematics & Statistics, South Dakota State University, Brookings, SD 57007
- 2 Office of Health Data and Research, Mississippi State Department of Health,570 East Woodrow Wilson, Jackson, MS 39215-1700
- 3 Pennington Biomedical Research Center, Louisiana State University System, 6400 Perkins Road,Baton Rouge, LA 70808
Open Journal of Statistics·Volume 02 (2012)·Pages 274–280·Published 27 June 2012·DOI10.4236/ojs.2012.23033
Copy link · social · email
Abstract
Empirical estimates of power and Type I error can be misleading if a statistical test does not perform at the stated rejection level under the null hypothesis. We employed the permutation test to control the empirical type I errors for zero-inflated exponential distributions. The simulation results indicated that the permutation test can be used effectively to control the type I errors near the nominal level even the sample sizes are small based on four statistical tests. Our results attest to the permutation test being a valuable adjunct to the current statistical methods for comparing distributions with underlying zero-inflated data structures.
KeywordsCentral Limit TheoremLikelihood Ratio TestModified Central Limit TheoremPermutation TestWald TestZero-Inflated Distribution
- D. M. Titterington, A. F. Smith and U. E. Makov, “Statistical Analysis of Finite Mixture Distributions,” John Wiley and Sons, New York, 1985.
- J. Lawless, “Negative Binomial and Mixed Poisson Regression,” Canadian Journal of Statistics, Vol. 15, No. 3, 1987, pp. 209-225. doi:10.2307/3314912
- D. C. Heibron, “Generalized Linear Models for Altered Zero Probability and Overdispersion in Count Data,” SIMS Technical Report No. 9, University of California, San Francisco, 1989.
- R. Schall, “Estimation in Generalized Linear Models with Random Effects,” Biometrika, Vol. 78, No. 4, 1991, pp. 719-727. doi:10.1093/biomet/78.4.719
- C. E. McCulloch, “Maximum Likelihood Algorithms for Generalized Linear Mixed Models,” Journal of American Statistical Association, Vol. 92, No. 437, 1997, pp. 162- 170.
- D. B. Hall, “Zero-Inflated Poisson and Binomial Regression with Random Effects: A Case Study,” Biometrics, Vol. 56, No. 4, 2000, pp. 1030-1039. doi:10.1111/j.0006-341X.2000.01030.x
- L. Zhang, J. Wu and W. D. Johnson, “Empirical Study of Six Tests for Equality of Populations with Zero-Inflated Continuous Distributions,” Communications in Statistics —Simulation and Computation, Vol. 39, No. 6, 2010, pp. 1196-1211. doi:10.1080/03610918.2010.489169
- G. Casella and R. L. Berger, “Statistical Inference,” Duxbury Inc., San Francisco, 2002.
- A. Wald, “Tests of Statistical Hypotheses Concerning Several Parameters When the Number of Observations Is Large,” Transactions in American Mathematical Society, Vol. 54, No. 3, 1943, pp. 426-482.
- E. S. Edgington, “Statistical Inference and Nonrandom Samples,” Psychological Series A, Vol. 66, No. 6, 1966, pp. 485-487. doi:10.1037/h0023916
- B. E. Wampold and N. L. Worsham, “Randomization Tests for Multiple Baseline Designs,” Behavioral Assessment, Vol. 8, 1986, pp. 135-143.
- R. C. Blair and W. Karniski, “An Alternative Method for Significance Testing of Waveform Difference Potentials,” Psychophysiology, Vol. 30, No. 5, 1993, pp. 518-524. doi:10.1111/j.1469-8986.1993.tb02075.x
- D. C. Adams and C. D. Anthony, “Using Randomization Techniques to Analyze Behavioural Data,” Animal Behaviour, Vol. 61, No. 1, 1996, pp. 733-738. doi:10.1006/anbe.2000.1576
- J. Ludbrook and H. Dudley, “Why Permutation Tests Are Superior to t and F Tests in Biomedical Research,” American Statistician Association, Vol. 52, No. 2, 1998, pp. 127-132.