Dynamic Seemingly Unrelated Cointegrating Regressions
提出动态似不相关回归（DSUR）估计量，用于同时估计多个协整回归，在方程间均衡误差相关时更有效，适用于时间序列长而截面小的面板协整估计，并应用于国民储蓄与投资关系、远期汇率预测偏差两个经典国际经济学问题。
We propose the parametric Dynamic Seemingly Unrelated Regression (DSUR) estimator for simultaneous estimation of multiple cointegrating regressions. DSUR is efficient when the equilibrium errors are correlated across equations and is applicable for panel cointegration estimation in environments where the cross section is small relative to the available time series. We study the asymptotic and small sample properties of the DSUR estimator for both heterogeneous and homogeneous cointegrating vectors. We then apply the method to analyse two long-standing problems in international economics. Our first application revisits the estimation of long-run correlations between national investment and national saving. Our second application revisits the question of whether the forward exchange rate is an unbiased predictor of the future spot rate. Copyright 2005, Wiley-Blackwell.