全球金融危機證實，過度且持續的信用創造及資產價格泡沫為觸發金融危機的重要因素。近來研究發現，評估金融情勢不僅有助預測未來總體經濟走勢，尚有助判定未來經濟下行風險。爰此，本文採用大量金融變數，並納入房地產市場與銀行信用等資訊，再以Doz, Giannone, and Reichlin (2011)兩階段動態因子模型，建構我國之金融情勢指數(financial conditions index, FCI)。實證結果顯示，FCI顯著領先實質面的經濟變數，且可提供有助經濟預測的額外資訊。當未來經濟成長的下行風險越高，FCI的領先性質將更形明顯；FCI對於經濟活動陷入衰退之預測能力表現佳，且納入房地產與信用變數之FCI表現可能更勝一籌。

In this paper, we develop Taiwan’s financial conditions index (FCI) using a wide range of financial indicators, especially encompassing real estate and credit data in order to study the relationship of financial conditions and economic forecasting. We employ the two-step dynamic factor model of Doz, Giannone, and Reichlin (2011) which makes estimation more accurate and efficient. The results indicate FCI not only performs better as a leading indicator of real economic variables, but also provides additional information for forecasting economic activity. Furthermore, the higher the future economic downside risks, the better the FCI serves as a leading indicator. FCI (specifically incorporating information on real estate and credit variables) appears to signal downside risks to economic growth well.