|
Activity Number:
|
667
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 4, 2016 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
|
Abstract #320198
|
|
|
Title:
|
Estimation of Heterogeneity for Multinomial Probit Models
|
|
Author(s):
|
Yixi Xu* and Qiang Liu and Xiao Wang
|
|
Companies:
|
and Purdue University and Purdue University
|
|
Keywords:
|
multinomial probit model ;
dictionary learning ;
EM algorithm ;
heterogeneity ;
random effect
|
|
Abstract:
|
In this paper, we study the multinomial probit model in which covariate effects are modeled semiparametrically. We incorporate heterogeneity across subjects by including the random effect for both parametric components and nonparametric components. In particular, novel techniques motivated from dictionary learning are developed for modeling individual nonparametric functions. Efficient algorithm is implemented to estimate the unknown parameters. The finite sample performance is illustrated by simulations and an application of brand choice study.
|
Authors who are presenting talks have a * after their name.
