本文應用多變量動態馬可夫轉換單因子模型,對台灣重要總體經濟變數如國內生產毛額、消費、投資以及出口進行估計,探討台灣經濟景氣循環轉折點之認定及預測。多變量動態馬可夫轉換單因子模型除了能夠刻劃景氣循環具有非對稱的特質外,也同時掌握景氣循環的另一重要特性:「重要總體經濟變數如消費、投資偏離時間趨勢的波動與國內生產毛額波動共同變動的現象」,這是單變量馬可夫轉換模型所無法做到的。實證結果發現多變量動態馬可夫轉換單因子模型確較單變量馬可夫轉換模型更有助於台灣景氣循環轉折點之認定與預測,尤其是在1990年代之後。
This paper builds upon the ideas proposed by Diebold and Rudebusch (1996) and estimates a multivariate dynamic Markov-switching factor model for a vector of macroeconomic variables. The approach captures both the idea of the business cycle as expressing co-movement in several macroeconomic variables as well as the asymmetric nature of business cycle phases. We transform the empirical models into state-space representation, and adopt Kim’s (1994) algorithm to implement the estimation. The empirical results suggest that the business chronologies identified by the multivariate Markov-switching factor model in terms of GDP, consumption and investment are more consistent with the CEPD-defined chronologies than those defined by the univariate Markov-switching models, especially for the post-1990 period.