Specification of mixed logit models assisted by an optimization framework

被引:28
作者
Paz, Alexander [1 ]
Arteaga, Cristian [2 ]
Cobos, Carlos [3 ]
机构
[1] Queensland Univ Technol, 2 George St, Brisbane, Qld 4000, Australia
[2] Univ Nevada, Transportat Res Ctr, 4505 Maryland Pkwy,POB 454007, Las Vegas, NV 89154 USA
[3] Univ Cauca, Informat Technol Res Grp GTI, Sect Tulcan Off 422, Popayan, Colombia
关键词
Mixed logit; Model specification; Optimization; Simulated annealing; Discrete outcome; PRINCIPAL COMPONENT ANALYSIS; VARIABLE SELECTION; LOGISTIC-REGRESSION; SUBSET-SELECTION; CALIBRATION; MACHINE;
D O I
10.1016/j.jocm.2019.01.001
中图分类号
F [经济];
学科分类号
02 ;
摘要
Mixed logit is a widely used discrete outcome model that requires for the analyst to make three important decisions that affect the quality of the model specification. These decisions are: 1) what variables are considered in the analysis, 2) which variables are to be modeled with random parameters, and 3) what density function do these parameters follow. The literature provides guidance; however, a strong statistical background and an ad hoc search process are required to obtain an adequate model specification. Knowledge and data about the problem context are required; also, the process is time consuming, and there is no certainty that the specified model is the best available. This paper proposes an algorithm to assist analysts in the search of an appropriate specification in terms of explanatory power and goodness of fit for mixed logit models. The specification includes the variables that should be considered as well as the random and deterministic parameters and their corresponding distributions. Three experiments were performed to test the effectiveness of the proposed algorithm. Comparison with existing model specifications for the same datasets were performed. The results suggest that the proposed algorithm can find adequate model specifications, thereby supporting the analyst in the modeling process.
引用
收藏
页码:50 / 60
页数:11
相关论文
共 43 条
[1]   Virtual multiphase flow metering using diverse neural network ensemble and adaptive simulated annealing [J].
AL-Qutami, Tareq Aziz ;
Ibrahim, Rosdiazli ;
Ismail, Idris ;
Ishak, Mohd Azmin .
EXPERT SYSTEMS WITH APPLICATIONS, 2018, 93 :72-85
[2]   Mathematical modeling and multi-start search simulated annealing for unequal-area facility layout problem [J].
Allahyari, Maral Zafar ;
Azab, Ahmed .
EXPERT SYSTEMS WITH APPLICATIONS, 2018, 91 :46-62
[3]  
[Anonymous], 1985, DISCRETE CHOICE ANAL
[4]   Computing the initial temperature of simulated annealing [J].
Ben-Ameur, W .
COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2004, 29 (03) :369-385
[5]   A MODIFICATION OF EFROYMSONS TECHNIQUE FOR STEPWISE REGRESSION ANALYSIS [J].
BREAUX, HJ .
COMMUNICATIONS OF THE ACM, 1968, 11 (08) :556-&
[6]  
Brownstone D, 1999, J ECONOMETRICS, V89, P109
[7]   A comparison of simulated annealing algorithms for variable selection in principal component analysis and discriminant analysis [J].
Brusco, Michael J. .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2014, 77 :38-53
[8]   Exact and approximate algorithms for variable selection in linear discriminant analysis [J].
Brusco, Michael J. ;
Steinley, Douglas .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2011, 55 (01) :123-131
[9]   Selection of multinomial logit models via association rules analysis [J].
Changpetch, Pannapa ;
Lin, Dennis K. J. .
WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL STATISTICS, 2013, 5 (01) :68-77
[10]  
Croissant Y., 2003, ESTIMATION MULTINOMI, V2-2, P73