Identifying the Predictors for Financial Crisis Using Gibbs Sampler

亞洲金融危機爆發後，是否危機能夠被事前預測以及何者因素爲適當的指標等議題再度吸引經濟學者的注意。在本文中，我們採用由George與McCulloch(1993)所發展出來的隨機變數選擇方法確認金融危機的指標。利用吉普抽樣的方法重覆模擬相關樣本，而出現頻率最高的樣本則被確認爲“最佳”模型。文中實證結果顯示，貨幣供給額M2年成長率與政府債務餘額佔國內生產毛額的比率，可作爲金融危機之重要指標。然而在文獻上經常被提到的指標，例如總外匯準備佔國內生產額比率與貿易帳赤字佔國內生產毛額比率，皆未被本文的分析方法所選入。本文的實證結果顯示貨幣與財政政策在亞洲金融危機中扮演非常關鍵的角色。

關鍵字

指標；金融危機；吉普抽樣

並列摘要

After the burst of Asian Financial Crisis, the issues of whether it is possible to predict a financial crisis and what are the appropriate predictors once again have attracted economists' attention. In this paper, we adopt the Stochastic Search Variable Selection (SSVS) developed by George and McCulloch (1993) to identify the crisis predictors. By using Gibbs sampler to simulate a correlated sample and ”best model” is identified as the one with highest frequency in the sample. Empirical results show that annual growth rate of money supply, M2, and the ratio of government debt to GDP are promising predictors for financial crisis. Somewhat surprisingly the frequently mention factors in recent literatures, such as ratio of total foreign reserve to GDP and the ratio of trade balance to GDP are not selected by our analysis. The empirical results imply that monetary and fiscal policy play a crucial role in exploring the Asian financial crisis.