消費者信心指數（CCI）是消費者對目前經濟狀況的滿意程度與對未來經濟預期的綜合性指標。由國外實證資料顯示，消費者信心指數是領先指標的重要構成項目，經濟研究機構常可藉由消費者信心指數所透露之事前訊息，來判定景氣循環的高峰和谷底，亦能預測未來景氣波動。故本計畫主要目的是利用不同計量預測模型，如VAR、Structural VAR、和Bayesian VAR等模型，檢視台灣消費者信心指數是否能有效預測景氣循環的高峰與谷底。此外，我們亦利用Granger causality探討台灣消費者信心指數是否為重要總體經濟變數的領先指標。最後，依據實證模型結果提供政府經濟預測與施政參考。 The consumer confidence index (CCI) measures overall consumer sentiments toward the short-term (6-month) future economic situation, and is regularly used to construct a lead indicator of economic trends. In this project we examine whether the consumer confidence index generally has good predictive power in identifying discrete turning points (peaks and troughs) in Taiwan’s business cycles. Firstly, compared with various forecasting models, including VAR, structural VAR, and Bayesian VAR models, we investigate the links between consumer confidence and economic activity. Secondly, we use the Granger causality tests to find out whether the consumer confidence index is a procyclical leading indicator in predicting business cycle activity. Finally, these empirical results will be informative to the government for drawing up macroeconomic policies. 研究期間 : 9801 ~ 9806
