PERMUTATION TESTS FOR LEAST ABSOLUTE DEVIATION REGRESSION

A permutation test based on proportionate reduction in sums of absolute deviations when passing from reduced to full parameter models is developed for testing hypotheses about least absolute deviation (LAD) estimates of conditional medians in linear regression models. Sampling simulations demonstrated that the permutation test on full model LAD estimates had greater relative power (1.06-1.43) than normal theory tests on least squares estimates for asymmetric, chi-square error distributions and symmetric, double exponential error distributions for models with one (n = 35 and n = 63) and three (n 63) independent variables. Relative power was .77 -.84 for normal error distributions. Power simulations demonstrated the low sensitivity of LAD estimates and permutation tests to outlier contamination and heteroscedasticity that was a linear function of X, and increased sensitivity to heteroscedasticity that was a function of X2 for simple regression models. Three permutation procedures for testing partial models in multiple regression were compared: permuting residuals from the reduced model, permuting residuals from the full model, and permuting the dependent variable. Permuting residuals from the reduced model maintained nominal error rates best under the null hypothesis for all error distributions and for correlated and uncorrelated independent variables. Power under the alternative hypotheses for this partial testing procedure was similar to full LAD regression model tests. Four example applications of LAD regression are provided.

[1]  Subhash C. NarulaI,et al.  The Minimum Sum of Absolute Errors Regression: A State of the Art Survey , 1982 .

[2]  Peter E. Kennedy Randomization Tests in Econometrics , 1995 .

[3]  G. Brown On Small-Sample Estimation , 1947 .

[4]  T. Nickelson,et al.  Influences of Upwelling, Ocean Temperature, and Smolt Abundance on Marine Survival of Coho Salmon (Oncorhynchus kisutch) in the Oregon Production Area , 1986 .

[5]  Joseph W. McKean,et al.  Coefficients of determination for least absolute deviation analysis , 1987 .

[6]  Don Edwards,et al.  Exact simulation-based inference: A survey, with additions , 1985 .

[7]  A. Allen Habitat suitability index models , 1986 .

[8]  J. Neter,et al.  Applied Linear Regression Models , 1983 .

[9]  R. Koenker,et al.  Robust Tests for Heteroscedasticity Based on Regression Quantiles , 1982 .

[10]  REGRESSION METHODOLOGY WITH GROSS OBSERVATION ERRORS IN THE EXPLANATORY VARIABLES , 1976 .

[11]  R. Carroll,et al.  Variance Function Estimation , 1987 .

[12]  Z. Ying,et al.  A resampling method based on pivotal estimating functions , 1994 .

[13]  Peter J. Huber,et al.  The place of the L1-norm in robust estimation , 1987 .

[14]  G. Baskerville Use of Logarithmic Regression in the Estimation of Plant Biomass , 1972 .

[15]  R. Koenker Confidence Intervals for Regression Quantiles , 1994 .

[16]  J. Gentle,et al.  Randomization and Monte Carlo Methods in Biology. , 1990 .

[17]  B. Manly Randomization, Bootstrap and Monte Carlo Methods in Biology , 2018 .

[18]  C. Braak,et al.  Permutation Versus Bootstrap Significance Tests in Multiple Regression and Anova , 1992 .

[19]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[20]  P. Mielke The application of multivariate permutation methods based on distance functions in the earth sciences , 1991 .

[21]  Mark F. Collins A Permutation Test for Planar Regression , 1987 .

[22]  Hannu Oja On Permutation Tests in Multiple Regression and Analysis of Covariance Problems , 1987 .

[23]  Terry E. Dielman,et al.  Bootstrapping in least absolute value regression: an application to hypothesis testing , 1988 .

[24]  Ian Barrodale,et al.  Algorithm 478: Solution of an Overdetermined System of Equations in the l1 Norm [F4] , 1974, Commun. ACM.

[25]  Daryl E. Carlson,et al.  A simple approximation of the sampling distribution of least absolute residuals regression estimates , 1977 .

[26]  R. Koenker,et al.  Asymptotic Theory of Least Absolute Error Regression , 1978 .

[27]  Joseph P. Romano Bootstrap and randomization tests of some nonparametric hypotheses , 1989 .

[28]  Randall M. Peterman,et al.  Application of Statistical Power Analysis to the Oregon Coho Salmon (Oncorhynchus kisutch) Problem , 1989 .

[29]  J. S. Maritz Distribution-Free Statistical Methods , 1981 .

[30]  Donald Ervin Knuth,et al.  The Art of Computer Programming, Volume II: Seminumerical Algorithms , 1970 .

[31]  Susan R. Wilson,et al.  Two guidelines for bootstrap hypothesis testing , 1991 .

[32]  William J. Welch,et al.  Construction of Permutation Tests , 1990 .

[33]  B. M. Brown,et al.  Distribution‐Free Methods in Regression1 , 1982 .

[34]  David Ruppert,et al.  Fitting heteroscedastic regression models , 1994 .

[35]  Frederick Mosteller,et al.  Data Analysis and Regression , 1978 .

[36]  H. G. Wilson,et al.  LEAST SQUARES VERSUS MINIMUM ABSOLUTE DEVIATIONS ESTIMATION IN LINEAR MODELS , 1978 .

[37]  David A. Freedman,et al.  A Nonstochastic Interpretation of Reported Significance Levels , 1983 .

[38]  Joyce Snell,et al.  6. Alternative Methods of Regression , 1996 .

[39]  D. W. Scott,et al.  The L 1 Method for Robust Nonparametric Regression , 1994 .

[40]  Roger Koenker,et al.  Tests of Linear Hypotheses and l[subscript]1 Estimation , 1982 .

[41]  Chih-Ling Tsai,et al.  Model selection for least absolute deviations regression in small samples , 1990 .