Permutation tests for univariate or multivariate analysis of variance and regression

Canadian Journal of Fisheries and Aquatic Sciences - Tập 58 Số 3 - Trang 626-639 - 2001
Marti J. Anderson

Tóm tắt

The most appropriate strategy to be used to create a permutation distribution for tests of individual terms in complex experimental designs is currently unclear. There are often many possibilities, including restricted permutation or permutation of some form of residuals. This paper provides a summary of recent empirical and theoretical results concerning available methods and gives recommendations for their use in univariate and multivariate applications. The focus of the paper is on complex designs in analysis of variance and multiple regression (i.e., linear models). The assumption of exchangeability required for a permutation test is assured by random allocation of treatments to units in experimental work. For observational data, exchangeability is tantamount to the assumption of independent and identically distributed errors under a null hypothesis. For partial regression, the method of permutation of residuals under a reduced model has been shown to provide the best test. For analysis of variance, one must first identify exchangeable units by considering expected mean squares. Then, one may generally produce either (i) an exact test by restricting permutations or (ii) an approximate test by permuting raw data or some form of residuals. The latter can provide a more powerful test in many situations.

Từ khóa


Tài liệu tham khảo

Anderson M.J., 2001, Aust. Ecol., 26, 32

Anderson M.J., 1999, J. Statist. Comput. Simul., 62, 271, 10.1080/00949659908811936

Anderson M.J., 2001, Austral. N.Z. J. Statist., 43, 75, 10.1111/1467-842X.00156

Boik R.J., 1987, Br. J. Math. Statist. Psychol., 40, 26, 10.1111/j.2044-8317.1987.tb00865.x

Brown B.M., 1982, Austral. J. Statist., 24, 318, 10.1111/j.1467-842X.1982.tb00837.x

Clarke K.R., 1993, Aust. J. Ecol., 18, 117, 10.1111/j.1442-9993.1993.tb00438.x

Fisher R.A., 1936, J. R. Anthropol. Inst. G.B. Irel., 66, 57

Fisher R.A., 1955, J. Roy. Statist. Soc., Ser. B, 17, 69

Freedman D., 1983, J. Busin. Econom. Statist., 1, 292

Gaston K.J., 1994, Philos. Trans. R. Soc. Lond. B, Biol. Sci., 335

Gonzalez L., 1998, Environmetrics, 9, 53, 10.1002/(SICI)1099-095X(199801/02)9:1<53::AID-ENV285>3.0.CO;2-#

Hayes A.F., 1996, Psychol. Methods, 1, 184, 10.1037/1082-989X.1.2.184

Hoeffding W., 1952, Ann. Math. Statist., 23, 169, 10.1214/aoms/1177729436

Hope A.C., 1968, J. Roy. Statist. Soc. Ser. B, 30, 582

Hurlbert S.H., 1984, Ecol. Monogr., 54, 187, 10.2307/1942661

Kempthorne O., 1955, J. Am. Stat. Assoc., 50, 946

Kempthorne O., 1966, J. Am. Stat. Assoc., 61, 11, 10.1080/01621459.1966.10502007

Kempthorne O., 1969, Biometrika, 56, 231, 10.1093/biomet/56.2.231

Kennedy P.E., 1996, Comm. Statist. Simulation Comput., 25, 923, 10.1080/03610919608813350

Neyman J., 1923, Roczniki Nauk Rolniczch, 10, 1

Pillar V.D.P., 1996, J. Veg. Sci., 7, 585, 10.2307/3236308

Pitman E.J.G., 1937, J. Roy. Statist. Soc. Ser. B, 4, 119

Pitman E.J.G., 1937, J. Roy. Statist. Soc. Ser. B, 4, 225

Pitman E.J.G., 1937, Biometrika, 29, 322

Romano J.P., 1988, Ann. Statist., 17, 141, 10.1214/aos/1176347007

Scheffé H., 1943, Ann. Math. Statist., 14, 305, 10.1214/aoms/1177731355

Still A.W., 1981, Br. J. Math. Statist. Psychol., 34, 243, 10.1111/j.2044-8317.1981.tb00634.x

Thông tin
Thông tin xuất bản

Canadian Journal of Fisheries and Aquatic Sciences

Tập 58 Số 3

626-639

Thông tin tác giả