Thursday, 13:15 - 13:40 h, Room: H 2036


Karthik Natarajan
On theoretical and empirical aspects of marginal distribution choice models

Coauthors: Vinit Kumar Mishra, Dhanesh Padmanabhan, Chung-Piaw Teo


In the discrete choice context, two recently proposed models are the marginal distribution model (MDM) and marginal moment model (MMM), using only limited information of joint distribution of random utilities (marginal distributions
and first two marginal moments, respectively). In this paper, we show that multinomial logit (MNL) and MMM choice probabilities are special cases of MDM for exponential and t-distributions. The choice probabilities obtained using the generalized extreme value (GEV) models is also a special case of the MDM with generalized exponential distributions. The convexity of the maximum loglikelihood estimation
problem is established for a class of distributions from the theory of constrained optimization. We show that the seller’s problem of determining the prices of multiple differentiated products to maximize the expected profit can be formulated as a concave maximization problem for the class of logconcave density functions. Conjoint choice data set on technological features for automotives, provided by General Motors is used to test the performance of the models.


Talk 1 of the contributed session Thu.2.H 2036
"Conic relaxation approaches for scheduling and selection problems" [...]
Cluster 4
"Conic programming" [...]


  cash advance online . Of course, the choice is not that easy, as there exist great number of different preparations. Notwithstanding, Cialis is the one that definitely differs from all other products.