×

Efficient and adaptive rank-based fits for linear models with skew-normal errors. (English) Zbl 1357.62244

Summary: The rank-based fit of a linear model is based on minimizing a norm. A score function needs to be selected for the fit and the proper choice leads to asymptotically efficient regression estimators, i.e., fits equivalent to the maximum likelihood estimators (mle). In this paper, we present the family of optimal scores functions for the skew-normal family of distributions. We show the easy computation of this rank-based fit using the \(R\) package Rfit. We present the results of a small simulation study comparing the rank-based estimators and the mles in terms of efficiency and validity over skew-normal and contaminated normal distributions. We also develop and present empirical results for a Hogg-type adaptive procedure for selecting among a family of these scores based on a robust initial fit.

MSC:

62J05 Linear regression; mixed models
62J07 Ridge regression; shrinkage estimators (Lasso)
62G35 Nonparametric robustness
62G05 Nonparametric estimation

Software:

sn; R; Rfit
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Abebe, A, McKean, JW: Weighted Wilcoxon estimators in Nonlinear Regression. Aust N Z J. Stat. 55, 401-420 (2013). · Zbl 1335.62113 · doi:10.1111/anzs.12046
[2] Abebe, A, McKean, JW, Kloke, JD, Bilgic, Y: Iterated Reweighted Rank-Based Estimates for GEE Models. Technical Report (2014). · Zbl 0628.62035
[3] Azzalini, A: A class of distributions which includes the normal ones. Scand. J. Stat. 12, 171-178 (1985). · Zbl 0581.62014
[4] Azzalini, A: R package sn: The skew-normal and skew-t distributions (version 0.4-18). Università di Padova, Italia (2013).
[5] Chang, W, McKean, J, Naranjo, J, Sheather, S: High-breakdown rank regression. J. Am. Stat. Assoc. 94, 205-219 (1999). · Zbl 1054.62570
[6] Hettmansperger, TP, McKean, JW: Robust Nonparametric Statistical Methods. 2nd Edition. Chapman-Hall, Boca Raton, FL (2011). · Zbl 1263.62048
[7] Hogg, RV, Fisher, DM, Randles, RH: A two-sample adaptive distribution-free test. J. Am. Stat. Assoc. 70, 656-661 (1975). · Zbl 0308.62030
[8] Hogg, RV, McKean, JW, Craig, AT: Introduction to Mathematical Statistics. 7th Ed. Pearson, Boston (2013).
[9] Huber, PJ: Robust Statistics. John Wiley & Son (1981). · Zbl 1388.62081
[10] Jaeckel, LA: Estimating regression coefficients by minimizing the dispersion of residuals. Ann. Math. Stat. 43, 1449-1458 (1972). · Zbl 0277.62049 · doi:10.1214/aoms/1177692377
[11] Jurečková, J: Nonparametric estimate of regression coefficients. Ann. Math. Stat. 42, 1328-1338 (1971). · Zbl 0225.62052 · doi:10.1214/aoms/1177693245
[12] Kloke, JD, McKean, JW: Rfit: Rank-based estimation for linear models. R J. 4, 57-64 (2012).
[13] Kloke, JD; McKean, JW, Small sample properties of JR estimators (2013), Alexandria, VA
[14] Kloke, JD, McKean, JW: Nonparametric statistical methods using R, Chapman-Hall, Boca Raton, FL (2014). · Zbl 0308.62030
[15] Kloke, JD, McKean, JW, Rashid, M: Rank-based estimation and associated inferences for linear models with cluster correlated errors. J. Am. Stat. Assoc. 104, 384-390 (2009). · Zbl 1388.62081 · doi:10.1198/jasa.2009.0116
[16] Koul, HL, Sievers, GL, McKean, JW: An estimator of the scale parameter for the rank analysis of linear models under general score functions. Scand. J. Stat. 14, 131-141 (1987). · Zbl 0628.62035
[17] McKean, J, Sheather, S: Diagnostic procedures. Wiley Interdiscip. S. Rev.: Comput tat. 1(2), 221-233 (2009). · doi:10.1002/wics.12
[18] McKean, J, Sievers, G: Rank scores suitable for analysis of linear models under asymmetric error distributions. Technometrics. 31, 207-218 (1989).
[19] Pourahmadi, M: Construction of skew-normal random variables: Are they linear combinations of normals and half-normals. J. Stat. Theory Appl. 3, 314-328 (2007).
[20] Shomrani, A: A comparison of different schemes for selecting and estimating score functions based on residuals. Ph.D. thesis, Western Michigan University, Department of Statistics (2003).
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.