.png){kind=icon}
.png){kind=icon}
Return
Linear double autoregression
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jeconom.2018.05.006.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper proposes the linear double autoregression, a conditional heteroscedastic model with a conditional mean structure but compatible with the quantile regression. The existence of a strictly stationary solution is discussed, for which a necessary and sufficient condition is established. A doubly weighted quantile regression estimation procedure is introduced, where the first set of weights ensures the asymptotic normality of the estimator and the second set improves its efficiency through balancing individual quantile regression estimators across multiple quantile levels. Bayesian information criteria are proposed for model selection, and two goodness-of-fit tests are constructed to check the adequacy of the fitted conditional mean and conditional scale structures. Simulation studies indicate that the proposed inference tools perform well in finite samples, and an empirical example illustrates the usefulness of the new model. (C) 2018 Elsevier B.V. All rights reserved.
Keywords:
Conditional quantile estimation
Goodness-of-fit test
Heavy tail
Nonlinear time series model
Stationary solution
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4
Papers:
5.2K
Citations:
3.0W