Use this URL to cite or link to this record in EThOS:
Title: Essays on econometric models of discrete choice for vertically differentiated alternatives in oligopoly markets
Author: Aristodemou, E.
ISNI:       0000 0004 8503 7370
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This thesis studies discrete choice demand models for vertically differentiated alternatives in oligopoly market settings and focuses on identification in discrete choice models motivated by demand. A cross-sectional discrete choice demand model is developed, where individuals choose the product that maximizes their utility, from a set of vertically differentiated alternatives offered by a set of horizontally differentiated brands. This allows the model to capture important features of individual substitution patterns when both kinds of differentiation are present. The unordered-ordered discrete nature of the two dimensions of the individual decision problem typically results in set identification of model parameters. The cross-sectional model is then extended to cover panel data settings, where individuals are observed for a number of periods. In many settings the choices individuals make in different periods are correlated such that current and future choices depend on past choices. For example, when individuals make purchasing decisions, they may form consumption habits and base their purchasing decisions on past firms' behaviour. Ignoring the dynamic behaviour in individuals' choices would result in inconsistent estimates of regression coefficients. This thesis studies discrete choice panel data models and extends the work on semiparametric identification of static binary panel data to static ordered panel data models, as well as to dynamic binary and ordered choice models. Set identification of the regression coefficients is achieved by differencing out the additively separable unobserved heterogeneity as in the case of linear panel data models.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available