欧洲杯买球网Recently, Prof. Xiaohui Liu from the School of Statistics and Data Science of our university, together with her collabor欧洲杯买球网ators Prof. Wei Long, Prof. Binduo Yang and Prof. Liang Peng, published a research paper named A Unified Inference for P欧洲杯决赛redictive Quantile Regression in the Journal of the American Statistical Association, the top international journal of s欧洲杯竞猜tatistics. The School of Statistics and Data Science and the Key Laboratory of Data Science for Finance and Economics of 欧洲杯买球网the University are the first joint units.

The paper proposes a new type of Unified Inference for Predictive Quantile Reg欧洲杯买球网ression based on data partitioning technique and stochastic weighted self-help method, which does not need to know wheth欧洲杯决赛er there is an intercept term in the model and is applicable to predictor variables with different persistence, and cond欧洲杯竞猜ucts a wide range of comparisons with similar existing methods based on Monte Carlo stochastic simulation and real data 欧洲杯赛事analysis. Based on Monte Carlo stochastic simulation and real data analysis, the proposed method is widely compared with 欧洲杯投注similar existing methods, and the superiority of the proposed method in terms of performance in limited samples is verified.

For more details, please 欧洲杯冠亚军refer to the original article.

Abstract:

The asymptotic behavior of quantile regression infe欧洲杯冠亚军rence becomes dramatically different when it involves a persistent predictor with zero or 欧洲杯比分nonzero intercept. Distinguishing various properties of a predictor is empirically challenging. In this article, we deve欧洲杯投注lop a unified predictability test for quantile regression regardless of the presence of intercept and persistence of a p欧洲杯比分redictor. The developed test is a novel combination of data splitting, weighted inference, and a random weighted bootstr欧洲杯决赛ap method. Monte Carlo simulations show that the new approach displays significantly better size and power performance t欧洲杯赛事han other competing methods in various scenarios, particularly when the predictive regress欧洲杯冠亚军or contains a nonzero intercept. In an empirical application, we revisit the quantile predictability of the monthly S&P 欧洲杯决赛500 returns between 1980 and 2019. Supplementary materials for this article are available online.

(By Wan Xinyue, Xie Yux欧洲杯赛事in, Lin Yang, Tao Chunhai, Zhang Mengyun)