Please use this identifier to cite or link to this item:
Title: Discrete Choice and Portfolio Optimization Under Limited Distributional Information
Keywords: Discrete Choice, Portfolio Optimization, Persistency Models, CVaR of Regret, Semiparametric Choice, Conjoint Choice, Packaging Design, SDP reduction
Issue Date: 15-Jun-2012
Citation: VINIT KUMAR MISHRA (2012-06-15). Discrete Choice and Portfolio Optimization Under Limited Distributional Information. ScholarBank@NUS Repository.
Abstract: The semiparametric Marginal Distribution Choice Models (MDM) are shown to give Generalized Extreme Value models and its popular special case Multinomial Logit Model (MNL). The MDM has nice structural properties for seller's pricing problem, and has computational advantages over simulation-based models such as Probit and mixed-logit for parameter estimation problem. The semiparametric Cross Moment Model (CMM) uses mean and covariance matrix of utilities and performs choice prediction using a Semidefinite program (SDP). The CMM nicely captures utility correlations and produces results comparable to Multinomial Probit, Nested Logit etc. when utilities are correlated. The results obtained by solving a packaging design problem using CMM and MNP show the strength of the CMM. The CMM SDP involves a large number of variables. Using SDP reduction techniques we propose a reduced but exact SDP for the CMM. A portfolio optimization problem is solved using this result with CVaR disutility under regret loss function. The results suggest that this model has high value under turbulent market conditions.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
MishraV.pdf5.79 MBAdobe PDF



Page view(s)

checked on Jun 7, 2019


checked on Jun 7, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.