scholarly journals Random-Coefficients Logit Demand Estimation with Zero-Valued Market Shares

2021 ◽  
Author(s):  
Jean-Pierre Dubé ◽  
Ali Hortaçsu ◽  
Joonhwi Joo
Keyword(s):  
Demand Estimation ◽  
Random Coefficients ◽  
Random Coefficient ◽  
Market Shares ◽  
Random Coefficient Logit

This paper provides a new estimator to address the zero-valued market shares in the BLP random-coefficient logit demand estimation.

An Introduction to Generalised Functional Form and Random Coefficients in Transport Modelling

1979 ◽  
Vol 11 (9) ◽  
pp. 1029-1037 ◽  
Author(s):  
L W Johnson
Keyword(s):  
Functional Form ◽  
Random Coefficients ◽  
Random Coefficient ◽  
Statistical Methodology ◽  
Transport Modelling ◽  
Regression Problems ◽  
Transportation Research

There are two developments in statistical methodology which have been examined extensively in the econometrics and statistics literature but have not been applied to a great extent in transportation research. These are the use of generalised functional form in regression problems, and random-coefficient techniques. The purposes of this paper are to present a brief survey of these areas, to examine the uses of these techniques on transport-related topics, and to suggest further applications of the procedures to transport. The exposition of these areas includes more intuitive than technical derivation, in order to make the paper as accessible as possible.

scholarly journals Monte Carlo Simulation in Random Coefficient Logit Models Involving Large Sums

2013 ◽  
Vol 1 (1) ◽  
pp. 85-108
Author(s):  
Zsolt Sándor
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Monte Carlo Methods ◽  
Mean Squared Error ◽  
Mixed Logit ◽  
Random Coefficient ◽  
Consumer Information ◽  
Logit Models ◽  
Quasi Monte Carlo ◽  
Random Coefficient Logit

Abstract We study Monte Carlo simulation in some recent versions of random coefficient logit models that contain large sums of expressions involving multivariate integrals. Such large sums occur in the random coefficient logit with demographic characteristics, the random coefficient logit with limited consumer information and the design of choice experiments for the panel mixed logit. We show that certain quasi-Monte Carlo methods, that is, so-called (t, m, s)-nets, provide improved performance over pseudo-Monte Carlo methods in terms of bias, standard deviation and root mean squared error.

Sign in / Sign up

Export Citation Format

Share Document