4.5 Article

Efficient Estimation for Models With Nonlinear Heteroscedasticity

Journal

JOURNAL OF BUSINESS & ECONOMIC STATISTICS
Volume 40, Issue 4, Pages 1498-1508

Publisher

TAYLOR & FRANCIS INC
DOI: 10.1080/07350015.2021.1933991

Keywords

Cramer-Rao bound; Efficient estimation; Nonlinear heteroscedasticity; Quantile regression

Ask authors/readers for more resources

This study introduces an efficient estimator by constrainedly weighting information across quantiles, which can eliminate the effect of preliminary estimator and achieve good estimation efficiency simultaneously. Compared to the Cramer-Rao lower bound, the relative efficiency loss of the new estimator has a conservative upper bound close to zero in practical situations. Monte Carlo studies show that the proposed method has substantial efficiency gain and better prediction performance in empirical applications to GDP and inflation rate modeling.
We study efficient estimation for models with nonlinear heteroscedasticity. In two-step quantile regression for heteroscedastic models, motivated by several undesirable issues caused by the preliminary estimator, we propose an efficient estimator by constrainedly weighting information across quantiles. When the weights are optimally chosen under certain constraints, the new estimator can simultaneously eliminate the effect of preliminary estimator as well as achieve good estimation efficiency. When compared to the Cramer-Rao lower bound, the relative efficiency loss of the new estimator has a conservative upper bound, regardless of the model design structure. The upper bound is close to zero for practical situations. In particular, the new estimator can asymptotically achieve the optimal Cramer-Rao lower bound if the noise has either a symmetric density or the asymmetric Laplace density. Monte Carlo studies show that the proposed method has substantial efficiency gain over existing ones. In an empirical application to GDP and inflation rate modeling, the proposed method has better prediction performance than existing methods.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.5
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available