A Unified Nonparametric Approach for Unbalanced Factorial Designs

Motivated by questions arising from the field of statistical genetics, we consider the problem of testing main, nested, and interaction effects in unbalanced factorial designs. Based on the concept of composite linear rank statistics, a new notion of weighted rank is proposed. Asymptotic normality of weighted linear rank statistics is established under mild conditions, and consistent estimators are developed for the corresponding limiting covariance structure. A unified framework to use weighted rank to construct test statistics for main, nested, and interaction effects in unbalanced factorial designs is established. The proposed tests statistics are applicable to unbalanced designs with arbitrary cell replicates greater than one per cell. The limiting distributions under both the null hypotheses and Pitman alternatives are derived. Monte Carlo simulations are conducted to confirm the validity and power of the proposed tests. Genetic datasets from a simulated backcross study are analyzed to demonstrate the application of the proposed tests in quantitative trait loci mapping.

[1]  J. L. Hodges,et al.  The Efficiency of Some Nonparametric Competitors of the t-Test , 1956 .

[2]  Ronald L. Iman,et al.  On some alternative procedures using ranks for the analysis of experimental designs , 1976 .

[3]  S. Rai.  Rank Analysis of Block Designs Having Different Cell Frequencies , 1987 .

[4]  C Brownie,et al.  A rank-based mixed model approach to multisite clinical trials. , 1992, Biometrics.

[5]  Larry P. Ammann,et al.  Efficacies of Rank-Transform Statistics in Two-Way Models with No Interaction , 1989 .

[6]  M. J. Prentice.  On the Problem of m Incomplete Rankings , 1979 .

[7]  G. A. Mack,et al.  A Friedman-Type Rank Test for Main Effects in a Two-Factor ANOVA , 1980 .

[8]  Edgar Brunner,et al.  Nonparametric Hypotheses and Rank Statistics for Unbalanced Factorial Designs , 1997 .

[9]  G. L. Thompson.  Amendments and Corrections: A Note on the Rank Transform for Interactions , 1993 .

[10]  L Kruglyak,et al.  A nonparametric approach for mapping quantitative trait loci. , 1995, Genetics.

[11]  Edgar Brunner,et al.  Nonparametric Methods for Stratified Two-Sample Designs with Application to Multiclinic Trials , 1995 .

[12]  G. A. Mack,et al.  On the Use of a Friedman-Type Statistic in Balanced and Unbalanced Block Designs , 1981 .

[13]  J. L. Hodges,et al.  Rank Methods for Combination of Independent Experiments in Analysis of Variance , 1962 .

[14]  H. Mansouri,et al.  Aligned rank transform tests in linear models , 1999 .

[15]  E. Lander,et al.  Mapping mendelian factors underlying quantitative traits using RFLP linkage maps. , 1989, Genetics.

[16]  J. Hájek,et al.  Asymptotic Normality of Simple Linear Rank Statistics Under Alternatives II , 1968 .

[17]  Michael G. Akritas,et al.  Limitations of the rank transform procedure: A study of repeated measures designs, Part II , 1993 .

[18]  Michael G. Akritas,et al.  The Rank Transform Method in Some Two-Factor Designs , 1990 .

[19]  P. vanElteren,et al.  A generalization of the method of M rankings : (Proceedings KNAW series A, 56(1953), nr 4, Indagationes mathematicae, 15(1953), p 358-369) , 1953 .

[20]  Jason P. Fine,et al.  On empirical likelihood for a semiparametric mixture model , 2002 .

[21]  W. J. Conover,et al.  The F Statistic in the Two-Way Layout With Rank–Score Transformed Data , 1984 .

[22]  Holger Dette,et al.  Rank procedures for the two-factor mixed model , 1992 .

[23]  K. S. Banerjee.  Generalized Inverse of Matrices and Its Applications , 1973 .

[24]  G. L. Thompson.  A note on the rank transform for interactions , 1991 .

[25]  Knut M. Wittkowski,et al.  Friedman-Type Statistics and Consistent Multiple Comparisons for Unbalanced Designs with Missing Data , 1988 .

[26]  R. Elston,et al.  The investigation of linkage between a quantitative trait and a marker locus , 1972, Behavior genetics.

[27]  Edgar Brunner,et al.  Rank statistics under dependent observations and applications to factorial designs , 1994 .

[28]  P. Sen,et al.  Nonparametric methods in multivariate analysis , 1974 .