我們使用隨機邊界模型(stochastic frontier model; Battese and Coelli, 1995)的公司技術無效率來評價由離散時間危險模型(discrete- time hazard model; Shumway, 2001)所得到的違約機率與標準普爾評等公司所公佈的長期發行者信用評等(long-term issuer credit rating; LTR)。為了研究LTR在技術無效率的效用,我們使用一個特殊設計矩陣來重新表示LTR的秩序等級。以1998年到2011年所有美國產業公司的月資料為樣本,實證結果顯示,標準普爾評等公司所公佈的LTR比離散時間危險模型所得到的違約機率,顯著地提供較多公司技術無效率的資訊內容(information content)。由於經濟效率是一個公司長期健康與前景的合理指標(Baek and Pag_an, 2002),我們結論標準普爾評等公司所公佈的LTR為一個好的公司信用風險代理者。
We use a stochastic frontier model with firm-specific technical inefficiency effects in a panel framework (Battese and Coelli, 1995) to assess the default probability (DP) based on the discrete-time hazard model (DHM; Shumway, 2001) and the long-term issuer credit rating (LTR) provided by Standard and Poor's Ratings Services (S&P). For studying the effect of S&P LTR on technical inefficiency, a special design matrix is used to represent the ordinal level of S&P LTR in the stochastic frontier model. For US industrial firms spanning the period 1998-2011 on a monthly basis, our empirical result shows that S&P LTR provides significantly more information content about firm's technical inefficiency than DP based on DHM. Combining the result with the fact that economic-based efficiency measures are reasonable indicators of the long-term health and prospects of firms (Baek and Pagán, 2002), we conclude that S&P LTR is a good credit risk proxy of firms.