×

A comparison of some MANOVA-type tests based on least distance. (English) Zbl 0929.62063

Summary: We consider some tests similar to those of classical multivariate analysis of variance but based on least distances rather than on least squares. The tests are compared. We also examine what sample size is sufficient to use approximations suggested by the asymptotic theory. The paper is a small-scale simulation study aimed at supplementing earlier asymptotic results. We conclude that tests which are known to have the same asymptotic properties, in fact behave quite differently for moderate sample sizes.

MSC:

62H15 Hypothesis testing in multivariate analysis
62J10 Analysis of variance and covariance (ANOVA)
65C60 Computational problems in statistics (MSC2010)
Full Text: DOI

References:

[1] Adamczyk K., Matematyka Stosowana 26 pp 75– (1993)
[2] DOI: 10.1016/0047-259X(88)90112-1 · Zbl 0649.62050 · doi:10.1016/0047-259X(88)90112-1
[3] Bai Z.D., Sankhyā 52 pp 166– (1990)
[4] Bai Z.D., Statistica Sinica 2 pp 237– (1992)
[5] DOI: 10.2307/2286611 · Zbl 0391.62046 · doi:10.2307/2286611
[6] DOI: 10.1007/BF00773354 · Zbl 0799.62061 · doi:10.1007/BF00773354
[7] DOI: 10.2307/2347150 · doi:10.2307/2347150
[8] Haldane J.B.S., Biometrika 25 pp 414– (1948)
[9] Koenker R., Statistical Data Analysis Based on L1 -norm and Related Methods pp 287– (1987)
[10] DOI: 10.2307/1913398 · Zbl 0497.62057 · doi:10.2307/1913398
[11] McKean J.W., Statistical Data Analysis Based on L1-norm and Related Methods pp 297– (1987)
[12] DOI: 10.1214/aos/1176348782 · Zbl 0786.62040 · doi:10.1214/aos/1176348782
[13] Niemiro W., Applicationes Math 22 pp 55– (1993)
[14] Niemiro W., Applications Math (1995)
[15] Rao C.R., Sankhyā 50 pp 289– (1988)
[16] Schrader R.M., In Statistical Data Analysis Based on L1-norm and Related Methods pp 307– (1987)
[17] Silverman B.W., Density Estimation for Statistics and Data Analysis (1986) · Zbl 0617.62042
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.