正态倒Gamma随机前沿模型的Bayesian推断

被引:3
作者
刘晓君 [1 ]
张世斌 [2 ]
机构
[1] 内蒙古大学数学科学学院
[2] 上海海事大学数学系
基金
上海市自然科学基金;
关键词
随机前沿模型; 倒Gamma分布; Bayesian推断; Gibbs抽样;
D O I
10.13299/j.cnki.amjcu.001790
中图分类号
O212.8 [贝叶斯统计];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
假设随机前沿模型的无效率项服从倒Gamma分布,利用Gibbs抽样方法对正态倒Gamma随机前沿模型参数进行Bayesian推断.导出了模型参数的后验条件分布,对中小型样本的模拟试验显示在最小后验均方误差准则下得到的参数估计值十分逼近真值.先验敏感性分析显示参数分布的后验均值相对于先验分布而言较为稳健.对电力公司实际数据分析显示正态倒Gamma随机前沿模型在拟合真实数据中有无效率项占总方差比重大的优点.
引用
收藏
页码:488 / 496
页数:9
相关论文
共 11 条
[1]  
高等数理统计[M]. 高等教育出版社 , 茆诗松, 2006
[2]   Maximum likelihood estimation of stochastic frontier models by the Fourier transform [J].
Tsionas, Efthymios G. .
JOURNAL OF ECONOMETRICS, 2012, 170 (01) :234-248
[3]   Bayesian and non-Bayesian analysis of gamma stochastic frontier models by Markov chain Monte Carlo methods [J].
Kozumi, H ;
Zhang, XY .
COMPUTATIONAL STATISTICS, 2005, 20 (04) :575-593
[4]   Estimation of technical inefficiencies with heterogeneous technologies [J].
Huang, HC .
JOURNAL OF PRODUCTIVITY ANALYSIS, 2004, 21 (03) :277-296
[5]  
Reconsidering heterogeneity in panel data estimators of the stochastic frontier model[J] . William Greene.Journal of Econometrics . 2004 (2)
[6]  
Simulated Likelihood Estimation of the Normal-Gamma Stochastic Frontier Function[J] . William H. Greene.Journal of Productivity Analysis . 2003 (2)
[7]   Full likelihood inference in normal-gamma stochastic frontier models [J].
Tsionas, EG .
JOURNAL OF PRODUCTIVITY ANALYSIS, 2000, 13 (03) :183-205
[8]   On the use of panel data in stochastic frontier models with improper priors [J].
Fernandez, C ;
Osiewalski, J ;
Steel, MFJ .
JOURNAL OF ECONOMETRICS, 1997, 79 (01) :169-193
[9]   Pitfalls of normal-gamma stochastic frontier models [J].
Ritter, C ;
Simar, L .
JOURNAL OF PRODUCTIVITY ANALYSIS, 1997, 8 (02) :167-182
[10]   SIMULATION OF TRUNCATED NORMAL VARIABLES [J].
ROBERT, CP .
STATISTICS AND COMPUTING, 1995, 5 (02) :121-125