Multinomial logit (MNL) models are widely used in marketing research to analyze choice data, but it is not generally recognized that the unit of the utility scale in a MNL model is inversely related to the error variance. This means that, for instance, parameters of two identical utility specifications estimated from different data sources with unequal variances will necessarily differ in magnitude, even if the true model parameters that generated the utilities are identical in both sets. Despite a growing number of papers that compare MNL coefficients, no examples of appropriate tests of the joint and separate hypotheses of scale and parameter equality in MNL models exist in the marketing literature. The purpose of this paper is to address the proper procedure for MNL parameter comparisons between different data sets and to propose a simple relative scaling test that can be implemented with standard MNL estimation software. Several examples are given to illustrate the approach.
Get full access to this article
View all access options for this article.
References
1.
AdamowiczWiktor, LouviereJordan, and WilliamsMichael (1992), “Combining Revealed and Stated Preference Methods to Value Environmental Amenities,” unpublished working paper, Department of Rural Economy, University of Alberta, Canada, (May).
2.
Ben-AkivaMoshe, and LermanSteve (1985), Discrete Choice Analysis: Theory and Application to Travel Demand.Cambridge, MA.: MIT Press.
3.
Ben-AkivaMoshe, and MorikawaTakayuki (1990), “Estimation of Travel Demand Models From Multiple Data Sources,” in Transportation and Traffic Theory, KoshiM., ed. New York: Elsevier, pp. 461–76.
4.
Ben-AkivaMoshe, MorikawaTakayuki, and ShiroishiFumiaki (1992), “Analysis of the Reliability of Preference Ranking Data,”Journal of Business Research, 24(2), 149–64.
5.
BerndtE., HallB., HallR., and HausmanJ. (1974), “Estimation and Inference in Nonlinear Structural Models,”Annals of Economic and Social Measurements, 3, 653–65.
6.
CurrimImran (1982), “Predictive Testing of Consumer Choice Models Not Subject to the Independence of Irrelevant Alternatives,”Journal of Marketing Research, 19(May), 208–22.
7.
GenschDavid (1985), “Empirically Testing a Disaggregate Choice Model for Segments,”Journal of Marketing Research, 22(November), 462–67.
8.
GuadagniPeter, and LittleJohn (1983), “A Logit Model of Brand Choice Calibrated on Scanner Data,”Marketing Science, 2(Summer), 203–38.
9.
HausmanJerry, and RuudPaul (1987), “Specifying and Testing Econometric Models for Rank-Ordered Data,”Journal of Econometrics, 34, 83–104.
10.
HorowitzJoel (1980), “Sources of Error and Uncertainty in Behavioral Travel Demand Models,” in New Horizons in Travel-Behavior Research, StopherP., MeyburgA., and BrogW., eds. Lexington, MA: Lexington Books, pp. 543–58.
11.
HorowitzJoel (1981), “Sampling, Specification and Data Errors in Probabilistic Discrete-Choice Models,” in Applied Discrete-Choice Modelling, HensherD., and JohnsonL., eds. New York: Wiley, pp. 417–35.
12.
HorowitzJoel, and LouviereJordan (1990), “The External Validity of Choice Models Based on Laboratory Choice Experiments,” in Spatial Choices and Processes, FischerM. M., NijkampP., and PapageorgiouY. Y., eds. Amsterdam: Elsevier North Holland, pp. 247–63.
13.
HorowitzJoel, and LouviereJordan (1992), “What Do Consideration Sets Really Measure?” unpublished working paper, Department of Marketing, Eccles School of Business, University of Utah, April.
14.
HorowitzJoel, and LouviereJordan (1993), “Testing Predicted Choices Against Observations in Probabilistic Discrete Choice Models,”Marketing Science, in press.
15.
JohnsonRichard, LouviereJordan, and OlsenDoug (1990), “Order and Practice Effects in Judgment and Choice Tasks,” presented at the Annual Meeting of the Marketing Science Institute, Champagne-Urbana, IL.
16.
JonesJ. Morgan, and LandwehrJane (1988), “Removing Heterogeneity Bias from Logit Model Estimation,”Marketing Science, 7(1), 41–59.
17.
KamakuraWagner, and RussellGary (1989), “A Probabilistic Choice Model for Market Segmentation and Elasticity Structure,”Journal of Marketing Research, 26(November), 379–90.
18.
LouviereJordan (1992), “Cross-Validating Different Conjoint Methods: Maybe It's Just a Question of Differing Variances?” working Paper, Eccles School of Business, University of Utah.
19.
LouviereJordan, and WoodworthGeorge (1983), “Design and Analysis of Simulated Consumer Choice or Allocation Experiments: An Approach Based on Aggregate Data,”Journal of Marketing Research, 20(November), 350–67.
20.
LouviereJordan, AndersonDon, and LouviereCathy (1991), “Bike Trail Choice Among Chicago Area Trail Users: Final Report to the Urban Forestry Project,” Department of Statistics, University of Wyoming.
21.
LouviereJordan, FoxMichael, and MooreWilliam (1992), “The Brand Constants Differ, but the Attribute Effects Are the Same: A Cross-Task Comparison of MNL Choice Models Estimated From Different Types of Stated Preference Data,” unpublished working paper, Department of Marketing, Eccles School of Business, University of Utah, (April).
22.
McFaddenDaniel (1981), “Econometric Models of Probabilistic Choice,” in Structural Analysis of Discrete Data With Econometric Applications, ManskiC., and McFaddenD., eds. Cambridge, MA: MIT Press, pp. 198–272.
23.
LouviereJordan (1986), “The Choice Theory Approach to Market Research,”Marketing Science, 5(4), 275–97.
24.
LouviereJordan (1987), “Regression-Based Specification Tests for the Multinomial Logit Model,”Journal of Econometrics, 34, 63–82.
25.
MorikawaTakayuki (1989), “Incorporating Stated Preference Data in Travel Demand Analysis,” unpublished PhD dissertation, Department of Civil Engineering, MIT, Cambridge, MA.
26.
SalomonIlan, and Ben-AkivaMoshe (1983), “The Use of the Life-Style Concept in Travel Demand Models,”Environment and Planning A, 15, 623–38.
27.
SeberGeorge, and WildChristopher (1989), Nonlinear Regression.New York: Wiley.
28.
SwaitJoffre, and Ben-AkivaMoshe (1986), “Constraints on Individual Travel Behavior in a Brazilian City,”Transportation Research Record1085, 75–85.
29.
SwaitJoffre, KozelValerie, BarrosRaimundo, and Ben-AkivaMoshe (1984), “A Model System of Individual Travel Behavior in a Brazilian City,”Transportation Policy and Decision-Making, 2, 451–80.
