臺灣博碩士論文加值系統

本文提出一個新的二階段模型來估計債務違約損失率(loss given default)，該
模型中的第一階段模型是使用羅吉斯迴歸模型(logistic regression model)來估計
違約損失率等於0 與大於0 的機率；第二階段模型是使用右尾延伸貝它模型
(right-tail extended beta model)來估計違約損失率介於(0,1) 的分配與等於1 的機
率。為了驗證這個新提議的二階段模型，我們從穆迪違約與回收率資料庫
(Moody’s Default and Recovery Database)中收集到4962 違約負債，實證結果顯
示，這個新的二階段模型產生準確違約損失率分配估計值，因此該模型可以用來
研究違約損失率分配。

We propose a new two-stage model to estimate the loss given default (LGD)
distribution. The first-stage model is the logistic regression model used to estimate
probabilities of LGD equal to 0 and larger 0, respectively. The second-stage model is
the right-tail extended beta model applied to generate the distribution of LGD between
(0,1) and probability of LGD equal to 1. To implement the newly proposed two-stage
model, we collect a sample of 4962 defaulted debts from Moody’s Default and
Recovery Database. The empirical results show that the newly proposed two-stage
model can generate the accurate LGD distribution estimate. Thus, it is useful for
studying the LGD distribution.

前言.......1
研究方法....3
資料來源....7
實驗結果....15
實證結果....19
參考文獻....21

Asarnow, E. and Edwards, D. (1995). Measuring loss on default bank loans: A 24-year study. Journal of Commercial Lending, 77, 11–23.

Altman, E. and Kalotay, EA.(2014). Ultimate recovery mixtures. Journal of Banking and Finance, 40,
116–129.

Bellotti, T. and Crook, J. (2012). Loss given default models incorporating macroeconomic variables
for credit cards. International Journal of Forecasting,
28,171–182.

Calabrese, R. (2014). Predicting bank loan recovery rates with mixed
continuous-discrete model. Applied Stochastic Models in Business and Industry,
30, 99–114.

Calabrese, R. and Zenga, M. (2010). Bank loan recovery rates: Measuring and nonparametric density
estimation. Journal of Banking and Finance, 34, 903–911.

Caselli, S., Gatti, S., and Querci, F. (2008). The sensitivity of the loss given default rate to
systematic risk: New empirical evidence on bank loans. Accepted by Journal of Financial Services
Research, 34, 1–34.

Duan, J.C. and Hwang, R.C., (2016). Predicting Recovery Rate at the Time of Corporate
Default. Available online at:
http://www.rmi.nus.edu.sg/duanjc/index_files/files/ConditionalRecoveryRate_Ju ne-26-2016.pdf

Li, P., Qi, M., Zhang, X., and Zhao, X. (2014) Further investigation of parametric loss given
default modeling. Available online at:
https://www.occ.gov/publications/publications-by-type/occ-working-papers/2012
-2009/wp2014-2.pdf

Loterman ,G., Brown, I., Martens, D., Mues, C., and Baesens, B. (2012).

Benchmarking regression algorithms for loss given default modeling.International Journal of Forecasting, 28, 161–170.

Siao, J.S., Hwang, R.C. and Chu, C.K., (2016). Predicting recovery rates using logistic
quantile regression with bounded outcomes. Journal of Quantitative Finance, 16, 777-792.

Sigrist, F. and Stahel, W.A., (2011). Using the censored gamma distribution for modeling fractional
response variables with an application to loss given default. ASTIN Bulletin, 41, 673–710.

Qi, M. and Zhao, X. (2011). Comparison of modeling methods for loss given default.Journal of Banking & Finance, 35,2842–2855.

Yashkir, O. and Yashkir, Y. (2013).Loss given default modeling: a comparative analysis. Journal of
Risk Model Validation 7, 25–59.