广西师范大学学报（自然科学版） ›› 2020, Vol. 38 ›› Issue (3): 59-69.doi: 10.16088/j.issn.1001-6600.2020.03.008

• • 上一篇    下一篇

负二项回归模型的重对数律和强相合性

杨晓伟1,2, 张军舰1*   

  1. 1.广西师范大学数学与统计学院,广西桂林541006;
    2.巢湖学院数学与统计学院,安徽合肥238000
  • 收稿日期:2019-03-03 出版日期:2020-05-25 发布日期:2020-06-11
  • 通讯作者: * 张军舰(1973—),男,河南内乡人,广西师范大学教授,博士。E-mail:jjzhang@gxnu.edu.cn
  • 基金资助:
    国家自然科学基金(11861017)

Law of Iterated Logarithm and Strong Consistency for Negative Binomial Regression Model

YANG Xiaowei1,2, ZHANG Junjian1*   

  1. 1. College of Mathematics and Statistics, Guangxi Normal University, Guilin Guangxi 541006, China;
    2. College of Mathematics and Statistics, Chaohu University, Hefei Anhui 238000, China
  • Received:2019-03-03 Online:2020-05-25 Published:2020-06-11

PDF (PC)

161

摘要： 本文构建负二项回归模型选择架构,研究在模型选择中对数似然函数的渐近性质及其与联系函数之间的依赖关系;推出模型参数最大似然估计的重对数律,并建立该类模型选择的强相合准则;证明在一些条件下,如果惩罚项阶数随模型的维数增加且介于O(lnln n)和O(n)之间,则由负最大对数似然和其惩罚项组成的模型选择准则几乎必然选择最简单的正确模型。

关键词: 重对数律, 负二项回归, 最大似然估计, 模型选择, 强相合性

Abstract: The structure of negative binomial regression model selection is constructed, and the asymptotic properties of logarithmic likelihood function and its dependence on relation function in model selection are studied. The law of iterated logarithm for maximum likelihood estimation of model parameters is derived, and the strong consistency criterion for selecting such models is established. It is proved that under some general conditions, if the order of penalty terms increases with the dimension of the model and is between O(lnln n) and O(n), the model selection criterion consisting of negative maximum logarithmic likelihood and penalty terms almost necessarily chooses the simplest correct model.

Key words: law of iterated logarithm, negative binomial regression, maximum likelihood estimation, model selection, strong consistency

中图分类号: 

  • O212.4
[1] 陈希孺. 广义线性模型的拟似然法[M].合肥: 中国科学技术大学出版社, 2011: 49-75.
[2] GEORGE E. Statistics in the 21st century[M].London: Chapman and Hall, 2002: 350-358.
[3] RAO C, WO Y. Model selection[M].Beachwood: Institute of Mathematical Statistics, 2001: 1-64.
[4] CAMERON A C, TRIVEDI P K. Regression-based tests for overdispersion in the Poisson model[J].Journal of Econometrics, 1990, 46(3):347-364. DOI: 10.1016/0304-4076(90)90014-K.
[5] QIAN G Q, FIELD C. Law of iterated logarithm and consistent model selection criterion in logistic regression[J]. Statistics and Probability Letters, 2002, 56(1): 101-112. DOI: 10.1016/S0167-7152(01)00191-2.
[6] AKAIKE H. Information theory and an extension of the maximum likelihood principle[C]//Proceedings of the Second International Symposium on Information Theory. Budapest: Akademiai Kiado,1973: 267-281.DOI: 10.1007/978-1-4612-0919- 5_38.
[7] SCHWARZ G. Estimating the dimension of a model[J].Annals of Statistics, 1978, 6(2): 461-464. DOI: 10.1214/aos/ 1176344136.
[8] RISSANEN J. Stochastic complexity in statistical inquiry[M].Singapore: World Scientific Publishing, 1989:123-133.
[9] TUTZ G. Regression for categorical data[M].London: Cambridge University Press, 2011: 143-162.
[10]WANG Z, MA S G, ZAPPITELLI M, et al. Penalized count data regression with application to hospital stay after pediatric cardiac surgery[J]. Statistical Methods in Medical Research, 2016, 26(5): 2685-2703.DOI:10.1177/0962280214530608.
[11]MALLOWS C. Some comments on Cp[J].Technometrics, 1973, 15(4): 661-675. DOI: 10.2307/1267380.
[12]QIAN G Q, KUNSCH H. Some notes on rissanen’s stochastic complexity[J].IEEE Transactions on Information Theory, 1998, 44(2): 782-786. DOI: 10.1109/18.661521.
[13]RISSANEN J. Fisher information and stochastic complexity[J].IEEE Transactions Information Theory, 1996, 42(1): 40-47. DOI: 10.1109/18.481776.
[14]RAO C, ZHAO L C. Linear representation of m-estimates in linear models[J].The Canadian Journal of Statistics, 1992, 20(4): 359-368. DOI: 10.2307/3315607.
[15]WU Y, ZEN M M. A strongly consistent information criterion for linear model selection based on M-estimation[J]. Probability Theory and Related Fields, 1999, 113(4): 599-625. DOI: 10.1007/s004400050219.
[16]QIAN G Q, WU Y H. Strong limit theorems on model selection in generalized linear regression with binomial responses[J].Statisticsa Sinica, 2006, 16(4): 1335-1365.
[17]RIGOLLET P. Kullback-Leibler aggregation and misspecified generalized linear models[J].The Annals of Statistics, 2012, 40(2): 639-665. DOI: 10.1214/11-AOS961.
[18]CHOW Y, TEICHER H. Probability Theory: independence, interchangeability, martingales[M].3rd ed.New York: Springer, 1997: 30-45.
[19]PETROV V. Limit theorems of probability theory: sequences of independent random variables[M].New York: Oxford University Press, 1995: 85-88.
[20]McCULLAGH P, NELDER J. Generalized linear models[M].2nd ed.London: Chapman and Hall, 1989: 90-112.
[21]BOHMAN H. Two inequalities for poisson distributions[J].Skandinavisk Aktuarietidskrift,1963,46(1): 47-52. DOI: 10.1080/03461238.1963.10404790.
[22]GEORGE E, McCULLOCK R. Approaches for bayesian variable selection[J].Statistica Sinica,1997, 7(2): 339-373.
[23]TIBSHIRANI R. Regression shrinkage and selection via the Lasso[J].Journal of the Royal Statistical Society,1996, 58(1): 267-288. DOI: 10.1.1.35.7574.
[1] 杨善朝, 梁丹. φ混合样本下频率插值密度估计的强相合性[J]. 广西师范大学学报（自然科学版）, 2012, 30(3): 16-21.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 唐国吉,赵婷,何登旭. 扰动广义混合变分不等式的可解性[J]. 广西师范大学学报（自然科学版）, 2018, 36(1): 76 -83 .
[2] 梁艳,周德雄,薛佳津,刘晓波,李俊,杨瑞云. 中药血散薯中非生物碱类化学成分研究[J]. 广西师范大学学报（自然科学版）, 2018, 36(1): 95 -98 .
[3] 许伦辉, 刘景柠, 朱群强, 王晴, 谢岩, 索圣超. 自动引导车路径偏差的控制研究[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 1 -6 .
[4] 邝先验, 吴赟, 曹韦华, 吴银凤. 城市混合非机动车流的元胞自动机仿真模型[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 7 -14 .
[5] 肖瑞杰, 刘野, 修晓明, 孔令江. 耦合腔光机械系统中两个机械振子的态交换[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 15 -19 .
[6] 黄慧琼, 覃运梅. 考虑驾驶员性格特性的超车模型研究[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 20 -26 .
[7] 袁乐平, 孙瑞山. 飞行冲突调配概率安全评估方法研究[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 27 -31 .
[8] 杨盼盼, 祝龙记, 操孟杰. 基于STM32的TSC型无功补偿控制系统的研究[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 32 -37 .
[9] 章美月. 关于电子束聚焦系统模型的一些新结果[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 38 -44 .
[10] 侯晓东, 蔡斌斌, 金炜东, 段旺旺. 基于证据距离和模糊熵的加权证据融合新方法[J]. 广西师范大学学报（自然科学版）, 2015, 33(1): 45 -51 .
版权所有 © 广西师范大学学报（自然科学版）编辑部
地址：广西桂林市三里店育才路15号　邮编：541004
电话：0773-5857325　E-mail: gxsdzkb@mailbox.gxnu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发