金融情勢指數（Financial Conditions Index）是根據眾多金融變數所編製而成的綜合指標，可以描述整體金融市場寬鬆與緊縮的狀況，也可作為主管機構調控貨幣政策的重要參考依據。不同於以往文獻，本文以兩步驟方法建構臺灣的金融情勢指數。首先，針對股票市場、貨幣市場與外匯市場，我們以Bai and Ng（2004）提出的PANIC模型，分別估計各市場中代表價、量變動的六個主要指標。其次，我們參考德意志銀行（Deutsche Bank）的金融情勢指數建構方式，透過景氣波動與各指數的線性迴歸分析，決定前一步驟所建構的六項指標分別在建構臺灣金融情勢指數時的組合權數。此時各項指標的組合權數所對應的正、負符號，均能符合總體經濟理論的認定限制。另一方面，根據Granger因果關係檢定與初步的樣本外預估表現評估，我們發現本文所建構的金融情勢指數具有領先大多數實質面總體變數的特性，此特性應可對更為深入與廣泛的經濟預測分析有所助益。

This paper develops a two-step approach for constructing Taiwan's financial conditions index (FCI). In the first step, by employing the Panel Analysis of Nonstationarity in Idiosyncratic and Common components (PANIC) proposed by Bai and Ng (2004), we obtain six common factors to capture the price and quantity fluctuations in the stock market, the money market and the foreign exchange market. In the second step, we construct the FCI based on these factors, with their weights determined by the method of Deutsche Bank, and find the resulting signs of these weights are consistent with the implications of macroeconomic theories. Moreover, the results of Granger causality tests and the pseudo out-of-sample experiments suggest that the FCI developed in this paper can lead the dynamics of many macroeconomic variables. This property may be helpful for improving economic forecasts.