Malhotra, Naresh K.
January 1987
Marketing Science;Winter87, Vol. 6 Issue 1, p98
Academic Journal
The modeling of consumer choice using procedures such as the logit, probit and other stochastic approaches is becoming increasingly popular in marketing. Recently, Dennis H. Gensch proposed a procedure for developing homogeneous segments for estimating disaggregate choice models using the logit approach. Unfortunately, the statistical test employed by Gensch to test the equality of logit coefficients estimated on two segments is incorrect. The note concludes with a recommendation to use the advocated test for testing the equality of parameters across segments when maximum likelihood estimation has been employed. Gensch has tested for poolability using the statistic, which he claims is distributed chi square. Gensch has made a technical error in that this statistic is not distributed chi square. Gensch performed his test on two samples of equal sizes. The issue of poolability may involve considerations other than the equality of estimated parameters across the segments. However, where the equality of logit parameters across segments is to be examined, the likelihood ratio test advocated in this note is strongly recommended over the test statistic proposed by Gensch.


Related Articles

  • A RESPONSE TO MALHOTRA'S COMMENT ON TESTING THE HOMOGENEITY OF SEGMENTS: REPLY. Ciensch, Dennis H. // Marketing Science;Winter87, Vol. 6 Issue 1, p100 

    The focus of Dennis H. Gensch's article is on the conceptualization of methodology to identify the "switchable" customers within a given segment or population. This methodology is independent of methods of segmenting a population. The empirical work on segmenting a population, reported in this...

  • ADVERTISING PLANNING: MATHEMATICAL MODELS IN ADVERTISING MEDIA PLANNING. Lodish, Leonard M. // Journal of Marketing Research (JMR);Feb1975, Vol. 12 Issue 1, p120 

    The article presents a review of the book "Advertising Planning: Mathematical Models in Advertising Media Planning," Dennis H. Gensch.

  • On The Overspecification of Multinomial and Nested Logit Models Due to Alternative Specific Constraints. Bierlaire, Michel; Lotan, Tsippy; Toint, Philippe // Transportation Science;Nov97, Vol. 31 Issue 4, p363 

    Discrete choice models as demand forecasting techniques have been used for transportation applications for more than thirty years. The multinomial and nested logit models are probably the most widely applied in this context. Alternative specific constants, although playing an important role in...

  • Characterizations through reliability measures from weight distributions. Navarro, J.; Del Aguila, Y.; Ruiz, J.M. // Statistical Papers;2001, Vol. 42 Issue 3, p395 

    Focuses on the derivation of the general characterizations of probability distributions from relationships between failure rate and mean residual life from the original distribution and associated weighted distribution. Formulation for the function inversion; Method of obtaining the...

  • The case against intermediate statistics. Lavenda, B. H.; Dunning-Davies, J. // Journal of Mathematical Physics;May89, Vol. 30 Issue 5, p1117 

    It is shown that intermediate statistics do not correspond to any physical process. The stationary probability distributions of intermediate statistics are not compatible with any mechanism which allows a variation between Fermi–Dirac and Bose–Einstein statistics. The binomial and...

  • Efficient Conjoint Choice Designs in the Presence of Respondent Heterogeneity. Yu, Jie; Goos, Peter; Vandebroek, Martina // Marketing Science;Jan/Feb2009, Vol. 28 Issue 1, p122 

    Random effects or mixed logit models are often used to model differences in consumer preferences. Data from choice experiments are needed to estimate the mean vector and the variances of the multivariate heterogeneity distribution involved. In this paper, an efficient algorithm is proposed to...

  • Two-sample rank tests based on exceeding observations. Eugenia Stoimenova // Applications of Mathematics;Aug2007, Vol. 52 Issue 4, p345 

    Abstract  Simple rank statistics are used to test that two samples come from the same distribution. Šid�k’s E-test (Apl. Mat. 22 (1977), 166–175) is based on the number of observations from one sample that exceed all observations from the other sample. A similar test...

  • Computing adjusted risk ratios and risk differences in Stata. Norton, Edward C.; Miller, Morgen M.; Kleinman, Lawrence C. // Stata Journal;2013, Vol. 13 Issue 3, p492 

    In this article, we explain how to calculate adjusted risk ratios and risk differences when reporting results from logit, probit, and related nonlinear models. Building on Stata's margins command, we create a new postestimation command, adjrr, that calculates adjusted risk ratios and adjusted...

  • Copula model generated by Dabrowska's association measure. D. Oakes; A. Wang // Biometrika;Jun2003, Vol. 90 Issue 2, p478 

    We propose a new archimedean copula model for bivariate survival data that is motivated by Dabrowska's (1988) measure of association. The model can represent negatively correlated or moderately positively correlated data but not highly positively correlated data. Local and global measures of...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics