Estimating Nonlinear Parameters in the Multinomial Logit Model Export

@article{citeulike:87217, abstract = {Multinomial logit models, especially those calibrated on scanner data, often use explanatory variables that are nonlinear functions of the parameters to be estimated. A common example is the smoothing constant in an exponentially weighted brand loyalty variable. Such parameters cannot be estimated directly using commercially available logit packages. We provide a simple iterative method for estimating nonlinear parameters at the same time as the usual linear coefficients. The procedure uses standard multinomial logit software and, in experience to date, converges rapidly. We prove that, under suitable conditions, the resulting parameter values are maximum likelihood estimates and show how to calculate asymptotic standard errors from normal computer output. Three applications illustrate the method in practice.}, author = {Fader, Peter S. and Lattin, James M. and Little, John D. C.}, citeulike-article-id = {87217}, citeulike-linkout-0 = {http://www.jstor.org/stable/184095}, journal = {Marketing Science}, keywords = {bryan}, number = {4}, pages = {372--385}, posted-at = {2007-02-14 23:15:36}, priority = {2}, title = {Estimating Nonlinear Parameters in the Multinomial Logit Model}, url = {http://www.jstor.org/stable/184095}, volume = {11}, year = {1992} }

TY - JOUR ID - citeulike:87217 L3 - citeulike-article-id:87217 N2 - Multinomial logit models, especially those calibrated on scanner data, often use explanatory variables that are nonlinear functions of the parameters to be estimated. A common example is the smoothing constant in an exponentially weighted brand loyalty variable. Such parameters cannot be estimated directly using commercially available logit packages. We provide a simple iterative method for estimating nonlinear parameters at the same time as the usual linear coefficients. The procedure uses standard multinomial logit software and, in experience to date, converges rapidly. We prove that, under suitable conditions, the resulting parameter values are maximum likelihood estimates and show how to calculate asymptotic standard errors from normal computer output. Three applications illustrate the method in practice. IS - 4 JF - Marketing Science EP - 385 TI - Estimating Nonlinear Parameters in the Multinomial Logit Model VL - 11 SP - 372 KW - bryan AU - Fader, Peter AU - Lattin, James AU - Little, John PY - 1992/// UR - http://www.jstor.org/stable/184095 ER -