Semiparametric versus Parametric Classification Models: An Application to Direct Marketing

In this paper we are concerned with estimation of a classification model using semiparametric and parametric methods. Benefits and limitations of semiparametric models in general, and of Manski's maximum score method in particular, are discussed. The maximum score method yields consistent estimates under very weak distributional assumptions. The maximum score method can very easily be used in situations where it is more serious to make one kind of classification error than another. In this paper, we use a so-called threshold-crossing model to discriminate between credit card holders and nonholders. The estimated parameters of the logit model differ significantly from the estimates of maximum score. Given an asymmetric loss function, maximum score performs better than the logit model.

[1]  R. Blundell,et al.  Modelling the Take-up of Means-tested Benefits: the Case of Housing Benefits in the United Kingdom , 1988 .

[2]  Daniel McFadden,et al.  A Comment on Discriminant Analysis "Versus" Logit Analysis , 1976 .

[3]  James M. Lattin,et al.  A Two-State Model of Purchase Incidence and Brand Choice , 1991 .

[4]  J. M. Jones,et al.  Removing Heterogeneity Bias from Logit Model Estimation , 1988 .

[5]  S. J. Press,et al.  Choosing between Logistic Regression and Discriminant Analysis , 1978 .

[6]  Jordan J. Louviere,et al.  Design and Analysis of Simulated Consumer Choice or Allocation Experiments: An Approach Based on Aggregate Data , 1983 .

[7]  P. K. Kannan,et al.  Modeling and Testing Structured Markets: A Nested Logit Approach , 1991 .

[8]  Anil K. Bera,et al.  Testing the Normality Assumption in Limited Dependent Variable Models , 1984 .

[9]  James L. Powell,et al.  Semiparametric Estimation of Selection Models: Some Empirical Results , 1990 .

[10]  D. Pollard,et al.  Cube Root Asymptotics , 1990 .

[11]  A. Deaton,et al.  COMPLETENESS , DISTRIBUTION RESTRICTIONS AND THE FORM OF AGGREGATE FUNCTIONS , 2022 .

[12]  Gordon G. Bechtel,et al.  Share-Ratio Estimation of the Nested Multinomial Logit Model , 1990 .

[13]  J. Horowitz.  A Smoothed Maximum Score Estimator for the Binary Response Model , 1992 .

[14]  C. Manski.  Semiparametric analysis of discrete response: Asymptotic properties of the maximum score estimator , 1985 .

[15]  R. Spady,et al.  AN EFFICIENT SEMIPARAMETRIC ESTIMATOR FOR BINARY RESPONSE MODELS , 1993 .

[16]  C. Manski.  MAXIMUM SCORE ESTIMATION OF THE STOCHASTIC UTILITY MODEL OF CHOICE , 1975 .

[17]  Dipak C. Jain,et al.  Effect of choice set size on choice probabilities: An extended logit model , 1989 .

[18]  B. Efron.  Bootstrap Methods: Another Look at the Jackknife , 1979 .

[19]  Richard R. Batsell,et al.  A Monté Carlo Comparison of Estimators for the Multinomial Logit Model , 1989 .

[20]  A. Ullah.  Nonparametric estimation and hypothesis testing in econometric models , 1988 .

[21]  Stephen Pudney,et al.  Modelling Individual Choice: The Econometrics of Corners, Kinks, and Holes , 1991 .

[22]  Peter Schmidt,et al.  The Theory and Practice of Econometrics , 1985 .

[23]  Pradeep K. Chintagunta,et al.  Investigating Heterogeneity in Brand Preferences in Logit Models for Panel Data , 1991 .

[24]  S. Cosslett.  DISTRIBUTION-FREE MAXIMUM LIKELIHOOD ESTIMATOR OF THE BINARY CHOICE MODEL1 , 1983 .

[25]  S. Sapra,et al.  A Connection between the Logit Model, Normal Discriminant Analysis, and Multivariate Normal Mixtures , 1991 .

[26]  P. Robinson.  Semiparametric econometrics: A survey , 1988 .

[27]  Naresh K. Malhotra,et al.  The Use of Linear Logit Models in Marketing Research , 1984 .

[28]  Charles F. Manski,et al.  Operational characteristics of maximum score estimation , 1986 .

[29]  Imran S. Currim.  Predictive Testing of Consumer Choice Models Not Subject to Independence of Irrelevant Alternatives , 1982 .