|
Activity Number:
|
180
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 1, 2016 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Business and Economic Statistics Section
|
|
Abstract #319200
|
View Presentation
|
|
Title:
|
Quantile Regression for Location-Scale Time Series Models with Conditional Heteroscedasticity
|
|
Author(s):
|
Jungsik Noh* and Sangyeol Lee
|
|
Companies:
|
The University of Texas Southwestern Medical Center and Seoul National University
|
|
Keywords:
|
ARMA-AGARCH models ;
asymptotic normality ;
CAViaR models ;
quantile regression
|
|
Abstract:
|
Quantile regression enables us to estimate conditional quantile functions and to explore the effect of covariates at each quantile level. This paper studies the quantile regression for general conditional location-scale time series models, which include ARMA models with asymmetric generalized autoregressive conditional heteroscedasticity (AGARCH) errors. The classical mean-variance models are reinterpreted as conditional location-scale models so that the quantile regression method can be naturally geared into the considered models. This paper provides a method for the joint estimation of model parameters in the conditional location and scale through the quantile regression. The consistency and asymptotic normality of the quantile regression estimator is established under mild conditions. The identifiability for the case of the ARMA-AGARCH model is also verified. This result facilitates a wider choice of models for value-at-risk forecasting. A simulation study and a real data example are reported in comparison with Gaussian quasi-maximum likelihood estimation.
|
Authors who are presenting talks have a * after their name.
