COINTEGRATION AND TREND-STATIONARITY IN MACROECONOMIC TIME-SERIES - EVIDENCE FROM THE LIKELIHOOD FUNCTION

被引:23
|
作者
DEJONG, DN
机构
[1] University of Pittsburgh, Pittsburgh
基金
美国国家科学基金会;
关键词
D O I
10.1016/0304-4076(92)90016-K
中图分类号
F [经济];
学科分类号
02 ;
摘要
A Bayesian approach to analyzing the co-integration inference is developed and applied to data deemed to be co-integrated by Engle and Granger (1987) and Campbell and Shiller (1987); the approach explicitly considers trend-stationary alternatives. It is shown that when trend-stationarity is given zero prior probability the data often appear to be co-integrated. When this prior is relaxed the data support trend-stationary representations.
引用
收藏
页码:347 / 370
页数:24
相关论文
共 50 条