A MONTE-CARLO COMPARISON OF TIME-VARYING PARAMETER AND MULTIPROCESS MIXTURE-MODELS IN THE PRESENCE OF STRUCTURAL SHIFTS AND OUTLIERS

• 成果类型: Note
• 署名作者: GAMBLE, JA; LESAGE, JP
• 署名单位: University System of Ohio; University of Toledo
• 刊物名称: REVIEW OF ECONOMICS AND STATISTICS
• ISSN/ISSBN: 0034-6535
• DOI: 10.2307/2109467
• 发表日期: 1993-08
• 页码: 515-519
• 关键词: money

摘要

This Monte Carlo study compares the performance of a recently proposed multiprocess mixture model and a more traditional random walk time-varying parameter (TVP) model in the face of structural shifts and outliers. The mixture model performs well and the latter model performs poorly. This finding is of general interest, since investigators often adopt random-walk TVP models to accommodate potential regime shifts in regression relationships. The findings suggest that the TVP estimation procedure is unlikely to find abrupt shifts, since the TVP parameter estimates are contaminated by the outliers and regime shifts.