Copula-based expectile regression: estimation and inference
提出一种基于Copula的期望分位数回归估计方法，将回归函数用Copula和边缘分布表示，证明渐近性质，模拟显示有限样本表现良好，实证用于分析成交量与汇率对股票收益的关系。
This article proposes a new approach to estimating the expectile regression function based on copulas. The main idea of this approach is to rewrite the expectile regression function in terms of a copula and marginal distributions. We show the asymptotic properties of our proposed estimator, for time series and iid settings, when the copula is estimated by maximizing the pseudo-log-likelihood and the margins are estimated nonparametrically. A Monte Carlo simulation study reveals that our estimator has good finite-sample properties for a variety of data-generating processes and different sample sizes. Finally, we provide two empirical applications to illustrate the practical relevance of the proposed methods. In these applications, we re-examined the relationship between volume and exchange rates on stock returns using copula-based expectile regressions. We found that the intercorrelation between two time series is a more important factor for improving the prediction than the autocorrelation in the time series.