张凌洁;苏美红;张海;

• 1:西北大学数学系

摘要(Abstract)：

主要针对损失函数为最小二乘LS(Least Squares)和最小绝对偏差LAD(Least Absolute Deviation)的凸组合形式,研究了观测数n和预测数p均趋于无穷大(lim n→∞p/n=κ,κ>0)时,高维稳健统计性质和高维罚稳健统计性质,得到了稳健估计和罚稳健估计的显示表达.结果显示这种凸组合损失函数的模型集成了LS和LAD损失的优点,同时消弱了它们的不足,具有优良的高维统计性质.

关键词(KeyWords)： 线性模型;高维;稳健估计;罚稳健估计;LS+LAD的凸组合

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金(11171272)

作者(Authors): 张凌洁;苏美红;张海;

参考文献(References)：

• [1]陈希孺,赵林城.线性模型中的M方法[M].上海:上海科学出版社,1996.
• [2]Huber P J.Robust estimation of a location parameter[J].Ann.Statist.,1964,35,73-101.
• [3]Huber P J.The 1972 Wald lecture.Robust statistics:A review[J].Ann.Statist.,1972,43:1041-1067.
• [4]Huber P J.Robust regression:asymptotics,conjectures and Monte Carlo[J].Ann.Statist.,1973,1:799-821.
• [5]Huber P J,Ronchetti E M.Robust Statistics[M].2nd ed.Hoboken,NJ:John Wiley and Sons Inc,2009.
• [6]Relles D.Robust Regression by Modifed Least Squares[C].New Haven:Ph.D.Thesis,Yale University,1968.
• [7]Yohai V J.Robust Estimation in the Linear Model[C].New Haven:Ph.D.Thesis,Yale University,1974.
• [8]Bickel P J.One-step Huber estimates in the linear model[J].J.Amer.Statist.Assoc.,1975,70:428-434.
• [9]Portnoy S.Asymptotic behavior of M-estimators of p regression parameters when p2/n is large[J].Ann.Statist.,1984,12:1298-1309.
• [10]Portnoy S.Asymptotic behavior of M estimators of p regression parameters when p2/n is large[J].Ann.Statist.,1985,13:1403-1417.
• [11]Le Cam L.On the assumptions used to prove asymptotic normality of maximum likelihood estimates[J].Ann.Statistics,1970,41:802-828.
• [12]Mammen E.Asymptotics with increasing dimension for robust regression with applications to the bootstrap[J].Ann.Statist.,1989,17:382-400.
• [13]Noureddine El Karoui,Derek Bean,Peter Bickel,et al.On robust regression with high-dimensional predictors[J].Proc.Natl.Acad.Sci.USA,2013,110(36):14557-14562.
• [14]Derek Bean,Peter Bickel,Noureddine El Karoui,et al.Penalized Robust Regression in High-Dimension[C].Berkeley:Technical Reports of Department of Statistics University of California,2011.
• [15]Derek Bean,Peter Bickel,Noureddine El Karoui,et al.Optimal Objective Function in High-Dimensional regression[C].Berkeley:Technical Reports of Department of Statistics University of California,2011.
• [16]Sculley D.Combined regression and ranking[J].Washington,DC,USA,K.D.D.,2010,10:25-28.
• [17]Wei Xiaoqiao.regression-based forecast combination methods[J].Romanian Journal for Economic Forecasting,2009(4):5-18.
• [18]Fan Yanqin.Asymptotic Normality of a combined regression estimator[J].Journal of Multivariate Analysis,1999,71:191-240.
• [19]Enrique Castillo,Carmen Castillo,Ali S Hadi,et al.Combined regression modles[J].Comput.Stat.,2009,24:37-66.
• [20]Rosasco E,De Vito A,Caponnetto M,et al.Are loss functions all the same?[J].Neural.Comput.,2003,16(5):1063-1076.
• [21]Akaike H.Information Theory and an Extension of the Mmaximum Likelihood Principle[D].Budapest:Akademiai Kiado,1973.
• [22]Schwarz G.Estimating the dimension of a model[J].Ann Stat.,1978,6:461-464.
• [23]Xu Zongben,Zhang Hai,Wang Yao,et al.L1/2Regularization[J].Sci.China Inf.Sci.,2010,53:1159-1169.
• [24]Zhang Hai,Liang Yong,Gou HaiLiang,et al.The essential ability of sparse reconstruction of diferent compressive sensing strategies[J].Sci.China Inf.Sci.,2012,55:2582-2589.
• [25]Xu Zongben,Chang Xiangyu,Xu Fengmin,et al.L1/2Regularization:a thresholding representation theory and a fast solver[J].Neural Networks and Learning Systems.2012,23(7):1013-1027.
• [26]Tibshirani R.Regression shrinkage and selection via the Lasso[J].Journal of the Royal Statistical Society Series,1996,58(1):267-288.
• [27]Mallows C L.Some comments on Cp[J].Technometrics,1973,15:661-675.
• [28]Peter Buhlmann Sara van de Geer.Statistics for High-Dimensional Data[M].New York:Springer,2010.
• [29]Fan J Q,Li R Z.Variable selection via nonconcave penalized likelihood and its oracle properties[J].Journal of the American Statistical Association,2001,31:1348-1360.