基于Gumbel Copula函数的多维Logit模型

李华民; 黄海军; 王惠文

基于Gumbel Copula函数的多维Logit模型

北京航空航天大学经济管理学院, 北京 100191

基金项目: 国家自然科学基金资助项目(70521001); 国家973基础研究计划资助项目(2006CB705503)

详细信息

作者简介:
李华民(1979-),男,河南平顶山人,博士生,li_lhm@sohu.com.

中图分类号: U 491
计量
- 文章访问数: 3720
- HTML全文浏览量: 147
- PDF下载量: 1422
- 被引次数: 0
出版历程
- 收稿日期: 2008-11-21
- 网络出版日期: 2009-12-31

Multinomial Logit model based on Gumbel Copula

School of Economics and Management, Beijing University of Aeronautics and Astronautics, Beijing 100191, China

摘要

摘要: 针对多维Logit模型中的独立同分布(IID, Identically & Independently Distributed)条件假设,提出了一种基于Copula函数的离散选择模型.利用Copula函数获得多元随机变量的联合分布函数以及Gumbel Copula函数的特性,得到了任意2个随机项之差的联合分布,它依然服从Logistic分布,形式上只比现有的分布函数多了一个倍参数.进一步将此结果推广至多维选择问题,得到了无需IID条件下一个方案被选中的概率,从而克服了多维Logit模型的应用障碍.
- Logit模型 /
- Gumbel Copula函数 /
- Logistic分布
Abstract: For the IID(identically & independently distributed) condition of the multinomial Logit model, which requires all random terms existing in the utilities of alternatives to be independent each other, a discrete choice model was proposed, based on the Copula function which can be used to derive the joint probability distribution of multi-random variables. The IID condition is weakened and the distribution of the difference between every two random terms is obtained using the Gumbel Copula function-s property. It is found that this distribution still follows the Logistic type but the difference must be multiplied by a parameter which can be estimated by the likelihood method from survey data. This result is then extended and employed in the discrete choice problem with more than two alternatives. The probability of choosing a specific alternative is rigorously formulated. The work surmounts the difficulty of applying the multinomial Logit model.
- Logit model /
- Gumbel Copula function /
- Logistic distribution

HTML全文

参考文献(1)

[1] McFadden D. Modeling the choice of residential location Karlqvist A,Snickars F,Weibull J.Spatial Interaction Theory and Planning Models.Amsterdam:North-Holland,1978:105-142 [2] 关宏志.非集计模型:交通行为分析的工具[M].北京:人民交通出版社,2004 Guan Hongzhi.Disaggregate mode:a tool of traffic behavior analysis[M].Beijing: China Communications Press,2004(in Chinese) [3] McFadden D,Train K E.Mixed MNL models of discrete response [J].Journal of Applied Econometrics,2000,15(5):447-470 [4] Hess S,Bierlaire M.Estimation of value of travel-time savings using mixed logit models [J].Transportation Research Part A,2005,39(3): 221-236 [5] Greene W H,Hensher D A.Accounting for heterogeneity in the variance of unobserved effects in mixed logit models [J].Transportation Research Part B,2006,40(1):75-92 [6] 陆化普,黄海军.交通规划理论研究前沿 [M].北京:清华大学出版社,2007:77-91 Lu Huapu,Huang Haijun.Theory research progress of transportation planning [M].Beijing: Tsinghua University Press,2007:77-91(in Chinese) [7] Train K E.Discrete choice methods with simulation[M].UK: Cambridge University Press,2003 [8] Nelsen R B.An introduction to Copulas[M].Second Edition.New York: Springer Verlag,2006 [9] Genest C,Rivest L P.A characterization of Gumbel-s family of extreme value distributions [J].Statist Probab Lett,1989,8(3): 207-211 [10] Hutchinson T P,Lai C D.Continuous bivariate distributions,emphasising applications [M].Adelaide: Rumsby Scientific Publishing,1990 [11] Joe H,Xu J J.The estimation method of inference functions for margins for multivariate models .Department of Statistics,University of British Columbia,Technical Report,166,1996 [12] 杨振海,程维虎.基于Logistic总体Ⅱ型截尾样本分布参数的近似极大似然估计[J].北京工业大学学报,2004,30(2):235-240 Yang Zhenhai,Cheng Weihu.Approximate maximum likelihood estimation based on Logistic distribution parameters of type Ⅱ censored samples [J].Journal of Beijing Polytechnic University,2004,30(2):235-240 (in Chinese)

施引文献

资源附件(0)

访问统计