臺灣博碩士論文加值系統

近幾年來，信用轉移矩陣已被大量投入至風險管理的領域，因為這些矩陣提供管理者一個簡單的評等轉變可能性(機率)，讓決策者能更有效的制定決策。本文將會介紹兩種適用於個體資料型態的估計方法，群組法和危險率法，以及一種適用於比例資料型態的估計方法，並且討論這些估計值的性質和準則，如：一致性、有效性等等。本文將這些估計方法實際應用到國家主權資料與公司財務資料上，資料方面考慮了19個國家的月資料，和50間美國製造業公司的季資料。另一方面，也會針對不同估計方法的結果做討論，比較表現之間的差異性。最後，我們會將國家主權資料估計結果與外匯投資組合做結合，以選取最佳投資組合，並在不同的投資策略下進行討論，比較各個投資策略表現的好壞。

Credit migration matrices are cardinal inputs to many risk management applications, and these matrices provide a succinct way of describing the evolution of credit ratings. This paper presents three methods to estimate transition matrices, including two methods by using individual data, the cohort approach and the hazard rate approach, and one method by using aggregate proportions data.

We will also show some properties of these estimators, for example, row sum condition, non-negativity condition, consistency, efficiency, and so on. We provide estimating results and their comparisons for both sovereign data and company data. In case studies, we consider monthly data of more than 19 countries and quarterly data of more than 50 companies. In addition, we discuss the merits of estimators by comparing their performances. Furthermore, we will apply the transition matrices to foreign exchange portfolio, and selecting the best foreign exchange portfolio by transition matrices.

中文摘要　v
英文摘要　vi
第一章　緒論　1
第二章　文獻回顧　3
　2-1　評等　3
　2-2　匯率投資組合　4
第三章　研究方法　5
　3-1　在個體資料下的轉移矩陣估計　5
　　3-1-1　群組法(Cohort approach)　5
　　3-1-2　危險率法(Hazard rate approach)　6
　3-2　在比例資料下的轉移矩陣估計　7
　　3-2-1　比例資料估計法　7
　3-3　轉移矩陣評分指標　9
　　3-3-1　評等移動速度(Rating moving-speed)　9
　　3-3-2　評等移動趨勢(Rating moving-trend)　10
　3-4　華德最小變異法(Ward’s method)　12
第四章　實證分析　13
　4-1　實證資料　13
　　4-1-1　國家主權資料選取　13
　　4-1-2　公司財務資料選取　14
　4-2　實證步驟　15
　　4-2-1　國家主權評等與公司評等　15
　　4-2-2　國家主權評等與公司評等之轉移矩陣比較　17
　　4-2-3　轉移矩陣估計對外匯投資組合結果之比較　18
　4-3　實證結果　19
　　4-3-1　評等結果　19
　　4-3-2　轉移矩陣之表現　21
　　4-3-3　投資組合之表現　30
第五章　結論　34
參考文獻　35
附錄一　公式推導　37
　1. 危險率法(Hazard rate approach)　37
　2. 比例資料估計法　37
附錄二　公司資訊　44
附錄三　實證分析圖　50
　國家主權評等結果　50
　公司評等結果　51
　各國匯率日資料　54
　各國匯率月報酬率　56
　轉移矩陣評分指標之比較　58
附錄四　實證分析表　61
　三種投資策略選取之資產　61

[1]Anderson, T. W. and Goodman, Leo A. (1957). Statistical Inference about Markov Chains. The Annals of Mathematical Statistics Vol. 28, 1, pp. 89-110.
[2]Baharumshah, A. Z., MacDonald, R., and Mohd, S. H. (2010). Exchange Rates in Singapore and Malaysia: Are They Driven by the Same Fundamentals Malaysian. Journal of Economic Studies, 47, pp. 123-41.
[3]Corte, P. D., Sarno, L., and Tsiakas, I. (2008). An Economic Evaluation of Empirical Exchange Rate Models. The Review of Financial Studies, 22, pp. 3491-530.
[4]Gapen, M., Gray, D., Cheng Hoon Lim, and Yingbin Xiao (2008). Measuring and Analyzing Sovereign Risk with Contingent Claims. IMF Staff Papers 55, pp. 109-148.
[5]Lando, D., Skodeberg, T. (2002). Analyzing ratings transitions and rating drift with continuous observations. Journal of Banking & Finance, 26 (2/3), pp. 423-444.
[6]Lee, T. C., Judge, G. G., and Zellner, A. (1970). Estimating the Parameters of the Markov Probability Model From Aggregate Time Series Data. North Holland.
[7]Jafry, Yusuf and Schuermann, Til (2004). Measurement, estimation and comparison of credit migration matrices. Journal of Banking & Finance 28, pp. 2603-39.
[8]Jones, Matthew T. (2005). Estimating Markov Transition Matrices Using Proportions Data：An Application to Credit Risk. IMF Working Paper.
[9]MacRae, Elisabeth Chase (1977). Estimation of Time-Varying Markov Processes with Aggregate Data. Econometrica Vol. 45, 1, pp. 183-198.
[10]Robert, Christian P., Ryden, Tobias, and Titterington, D. M. (2000). Bayesian Inference in Hidden Markov Models through the Reversible Jump Markov Chain Monte Carlo Method. Journal of the Royal Statistical Society Series B (Statistical Methodology), 62, pp. 57-75.
[11]Wilson, Thomas C. (1997). Portfolio Credit Risk.
[12]吳靖東 (2011)。總體經濟因素對兩岸匯率變動的影響。東吳經濟商學報 76，pp. 99-112。
[13]徐千婷 (2006)。匯率與總體經濟變數之關係：台灣實證分析。中央銀行季刊 28，pp. 13-42。
[14]謝劍平 (2012)。固定收益證券：債券市場與投資策略(三版)。台北：智勝出版社。