删失回归模型中一个LASSO型变量选择和估计方法(英文)

A LASSO-Type Approach to Variable Selection and Estimation for Censored Regression Model

删失回归模型是一种很重要的模型,它在计量经济学中有着广泛的应用.然而,它的变量选择问题在现今的参考文献中研究的比较少.本文提出了一个LASSO型变量选择和估计方法,称之为多样化惩罚L1限制方法,简称为DPLC.另外,我们给出了非0回归系数估计的大样本渐近性质.最后,大量的模拟研究表明了DPLC方法和一般的最优子集选择方法在变量选择和估计方面有着相同的能力.

Censored regression （＂Tobit＂） model is one of important widely used in econometrics. However, studies for variable selection problem in censored regression model are rare at the present references. In this paper, for censored regression model we propose a LASSO-type approach, diverse penalty L1 constraint method （DPLC）, to select variables and estimate the corresponding coefficients. Furthermore, we obtain the asymptotic properties of nonzero elements＇ estimation of regression coefficient. Finally, extensive simulation studies show that DPLC method almost possesses the same performance of selecting variables and estimation as generally best subset selection method （GBSS）.