Browse > Article

Development of Scaled Explosion Logit Model Considering Reliability of Ranking Data  

Kim, Kang-Soo (교통개발연구원)
Cho, Hye-Jin (한국건설기술연구원)
Publication Information
Journal of Korean Society of Transportation / v.22, no.6, 2004 , pp. 197-206 More about this Journal
Abstract
In ranking data, respondents are required to rank a number of alternatives in order of their preferences and an exploded logit model is generally used. It assumes that each rank contains the same amount of random noise. This study investigates the reliability of ranking data and identifies whether there are different decision rules at each rank stage. The results show that there were differences in the amount of unexplained variation in different ranking stage. A single scaling parameter could not explain the difference of variations of individual coefficients between two ranking data average difference of variations. This paper also investigated the optimal explosion depth in the exploded logit model by using the suggested scaling approach. The scaling approach should be based on particular variables which have different variances rather than based on the whole data set. The empirical analysis show that an explosion depth of 2 is appropriate after scaling the second rank data set, while an explosion including the third rank is inappropriate even though the third rank data set is scaled.
Keywords
SP; Ranking Data; Exploded Logit; Reliability; Scaling;
Citations & Related Records
Times Cited By KSCI : 1  (Citation Analysis)
연도 인용수 순위
1 Beggs, S.S. Cardell, and J. Hausman(1981), 'Assessing the Potential Demand for Electric Cars', Journal fo Econometrics, Vol.16, pp.1-19   DOI   ScienceOn
2 Chapman R. G. and Staelin, R. G. (1982) 'Exploring rank ordered choice set data within the stochastic utility model', Journal of Marketing Research Vol.19, pp. 288-301   DOI   ScienceOn
3 Ben-Akiva, M. and Lerman, D.(1985) 'Discrete Choice Analysis: Theory and Application to Travel Demand', MIT Press, Cambridge
4 Daly, A. J. (1992) 'ALOGIT 3.2 User's Guide, Hague Consulting Group', The Hague
5 Barbera, S. and P.K. Pattanaik(1986), 'Falmagne and the Rationalizability of Stochastic Choice in Terms of Random Orderings', Econometrica, Vol.54, No.3, pp.707-715   DOI   ScienceOn
6 강혜영(2000), '선호의식자료를 이용한 주차선택행태모형개발에 관한 연구', 한양대학교 환경대학원 석사학위 논문
7 김강수. 조혜진(2003), ‘Error Component 방법을 이용한 RP.SP 결합모형 개발’, 대한교통학회지, 제21권 제2호, 대한교통학회, pp.119-130
8 Krantz, D.H. and Tversky(1971), 'A Conjoint Measurement Analysis of composition rules in psychology', Psychological Review, (78) pp.151-169   DOI
9 김강수(2001), 'Stated Preference 조사설계 및 분석방법론에 대한 연구(2단계)', 최종보고서, 교통개발연구원
10 진교남(1997), ‘교통수단선택모형의 추정에 이용되는 선호의식 자료의 유효성에 관한 연구’, 서울대학교 대학원 박사논문
11 Luce, R.D. and Suppes, P.(1965) 'Preference, Utility, and Subjective Probability', Handbook of Mathematical Psychology 3, R.D. Luce(ed), John Wiley & Sons, Inc., New York, pp.294-410
12 Bradley, M. A. and Daly, A. J.(1994) 'Uses of the Logit Scaling Approach in Stated Preference Analysis', Transportation Vol.21, pp.167-184   DOI
13 Kim, K. S. (1998) 'A Behavioural Approach to Freight Transport Model Choice', Institute for Transport Studies, University of Leeds, Ph.D Thesis