二元选择分位数回归模型的贝叶斯估计方法及模拟研究
|
摘要
文章针对二元选择分位数回归模型的贝叶斯估计方法进行探索性研究。通过模拟实验比较分析了不同先验设定和不同抽样算法下贝叶斯二元选择分位数回归估计量的统计性质,以及贝叶斯方法与频率学派方法对二元选择分位数模型进行估计的优劣。结果表明,对二元选择分位数回归模型进行贝叶斯估计具有一定的优势,尤其是小样本下,估计效果更佳;而且采用Gibbs抽样得到的估计量精度更高,统计推断更准确,尤其是在高分位数下Gibbs抽样的优势更明显。
This paper makes an exploratory study on Bayesian estimation of binary selection quantile regression model. By simulation experiments the paper comparatively analyzes the statistical properties of Bayesian binary selection quantile regression estimator under different priori settings and different sampling algorithms, as well as the merits and demerits of Bayes method and frequency school method to estimate the binary choice quantile model. The results show that the Bayesian estimation of binary selection quantile regression model has certain advantages, especially in the case of small samples, and the estimation effect is better. Moreover, the estimator precision obtained by Gibbs sampling is higher and the statistical inference more accurate; especially in the case of high score, Gibbs sampling has more obvious advantages.
This paper makes an exploratory study on Bayesian estimation of binary selection quantile regression model. By simulation experiments the paper comparatively analyzes the statistical properties of Bayesian binary selection quantile regression estimator under different priori settings and different sampling algorithms, as well as the merits and demerits of Bayes method and frequency school method to estimate the binary choice quantile model. The results show that the Bayesian estimation of binary selection quantile regression model has certain advantages, especially in the case of small samples, and the estimation effect is better. Moreover, the estimator precision obtained by Gibbs sampling is higher and the statistical inference more accurate; especially in the case of high score, Gibbs sampling has more obvious advantages.
引文
[1]Koenker R,Bassett G S,Regression Quantiles[J].Econometrica,1978,(46).
[2]Yu K,Moyeed R A.Bayesian Quantile Regression[J].Statistics and Probability Letters,2001,(54).
[3]Manski C F.Maximum Score Estimation of the Stochastic Utility Model of Choice[J].Journal of Econometrics,1975,3(3).
[4]Manski C F.Semiparametric Analysis of Discrete Response:Asymptotic Properties of the Maximum Score Estimator[J].Journal of Econometrics,1985,27(3).
[5]Koenker R,Hallock F.Quantile Regression[J].Journal of Economic Perspectives,2001,15(4).
[6]Kordas G.Smoothed Binary Regression Quantiles[J].Journal of Applied Econometrics,2006,21(3).
[7]Skouras S.An Algorithm for Computing Estimators that Optimize Step Functions[J].Computational Statistics and Data Analysis,2003,42(3).
[8]Florios K,Skouras S.Exact Computation of Max Weighted Score Estimators[J].Journal of Econometrics,2008,146(1).
[9]Powell J.Censored Regression Quantiles[J].Journal of Econometrics,1986,32(1).
[10]Kim J,Pollard D.Cube Root Asymptotics[J].Annals of Statistics,1990,18(1).
[11]Delgado M A,Rodriguez-Poo J M,Wolf M.Subsam Pling Inferences in Cube Root Asymptotics With an Application to Manski’s Maximum Score Estimator[J].Economic Letters,2001,73(2).
[12]Abrevaya J,Huang J.On the Bootstrap of the Maximum Score Estimator[J].Econometrica,2005,73(4).
[13]Benoit D F,Van den Poel D.Binary Quantile Regression:A Bayesian Approach Based on the Asymmetric Laplace Distribution[J].Journal of Applied Econometrics,2012,7(7).
[14]Kottas A,Krnjajic M.Bayesian Semiparametric Modeling in Quantile Regression[J].Scandinavian Journal of Statistics,2009,(36).
[2]Yu K,Moyeed R A.Bayesian Quantile Regression[J].Statistics and Probability Letters,2001,(54).
[3]Manski C F.Maximum Score Estimation of the Stochastic Utility Model of Choice[J].Journal of Econometrics,1975,3(3).
[4]Manski C F.Semiparametric Analysis of Discrete Response:Asymptotic Properties of the Maximum Score Estimator[J].Journal of Econometrics,1985,27(3).
[5]Koenker R,Hallock F.Quantile Regression[J].Journal of Economic Perspectives,2001,15(4).
[6]Kordas G.Smoothed Binary Regression Quantiles[J].Journal of Applied Econometrics,2006,21(3).
[7]Skouras S.An Algorithm for Computing Estimators that Optimize Step Functions[J].Computational Statistics and Data Analysis,2003,42(3).
[8]Florios K,Skouras S.Exact Computation of Max Weighted Score Estimators[J].Journal of Econometrics,2008,146(1).
[9]Powell J.Censored Regression Quantiles[J].Journal of Econometrics,1986,32(1).
[10]Kim J,Pollard D.Cube Root Asymptotics[J].Annals of Statistics,1990,18(1).
[11]Delgado M A,Rodriguez-Poo J M,Wolf M.Subsam Pling Inferences in Cube Root Asymptotics With an Application to Manski’s Maximum Score Estimator[J].Economic Letters,2001,73(2).
[12]Abrevaya J,Huang J.On the Bootstrap of the Maximum Score Estimator[J].Econometrica,2005,73(4).
[13]Benoit D F,Van den Poel D.Binary Quantile Regression:A Bayesian Approach Based on the Asymmetric Laplace Distribution[J].Journal of Applied Econometrics,2012,7(7).
[14]Kottas A,Krnjajic M.Bayesian Semiparametric Modeling in Quantile Regression[J].Scandinavian Journal of Statistics,2009,(36).