臺灣博碩士論文加值系統

本論文主要是探討市場風險值之風險 — 其風險來源特性 , 風險值應用, 新巴塞爾資本協定對金融機構之影響, 暨在學術及金融機構實務中較廣泛討論風險值理論模型. 臺灣至民國91年4月已成立14家金融控股公司, 新巴塞爾資本協定及市場風險之新規定對新成立金融控股公司有顯著之影響 - 財政部已宣示風險控管是未來對金融控股公司金融撿查重點.
因為交易資料之隱密行及無法自公開資料源取得, 一般有關風險值之研究侷限於模型研究及比較, 或作單一資產之波動分析. 直到2002年6月 (Journal of Finance) 才第一次以6家美國大銀行 2年內部所計每日損益資料及依銀行年內部模型產出每日風險值預測實證研究與GARCH 模型比較風險值預準確性, 實證研究結果是GARCH 模型較銀行內部模型更精準測預風險值.
本研究中並以二組資產組合 (分別模擬2家金控公司交易) ,分別以5種風險值預測模型, GRACH(1,1)-AR(1), GARCHM(1,1), RiskMetricTM, Historical Simulation and Monte Carlo Simulation,於 95% 及99% 性賴區間 (一日時間) 實證研究 , 結果是於99% 性賴區間 GARCH(1,1)-AR(1) 及 Historical Simulation 二者表現最佳僅有一次預測值操出範圍 , 於95% 性賴區間 Historical Simulation 表現最佳.

THESIS ABSTRCT
GRADUATE INSTITUTE OF BUSINESS ADMINISTRATION
NATIONAL TAIWAN UNIVSRITY
NAME : CHIEN-YIN, WANG MONTH/YEAR : JUNE, 2003.
ADVISER : PROFESSOR YONG-CHERN SU
MARKET RISK VAR MODELS FOR FINANCIAL HOLDING COMPANY
This study is to discuss the conceptual, the application, the Basle II requirements and the commonly adopted methodologies & models, of market risk VaR, both in empirical research and in practice. Then we compare the performance of these discussed VaR prediction models by two portfolios. Most of the VaR analyses in the public domain have been limited to comparing modeling approaches and implementation procedures using illustrative portfolios because of the proprietary mature of real data results from the financial organizations. There was only one research paper (Berkowitz & O’Brien (2002)) examining the statistical accuracy of the VaR forecasts and provide direct evidence on the performance of bank of VaR models. The key important findings are that the GARCH model of P&L generally provides for lower VaRs and is better at predicting changes in volatility.
Due to the lack of availability of the trading data & VaR predicted data from financial institutions, we simulate two financial holding companies’ portfolios with asset classes of foreign exchange, bond and securities. Then we compute VaR predictions and compare the of GARCH(1,1)-AR(1), GRACHM(1,1), RiskMetricsTM, historical simulation and Monte Carlo Simulation , of the two simulated portfolios. We test the VaR prediction in an one-day horizon at 95 and 99 percent confidence level. Conclusion from the testing is on 99 percent confidence level both historical simulation and GARCH(1,1)-AR(1) are equally the best to predict the VaR for both portfolio A & B — i.e. only depicting one failure in the period of 217 observations. While on 95 percent testing historical simulation provides a best result and then follows GARCH (1,1)-AR (1).
The contribution of this paper is to provide clear walkthrough in market risk; sources, regulations, models available and performance results for two different asset mixed portfolios.

Chapter 1 Introduction 1
1.1 Motives and Purposes 1
1.2 Framework of the Thesis 6
Chapter 2 Literature Review 8
2.1 Types of Financial Risks 8
2.1.1 Market Risk 10
2.1.2 Credit Risk 10
2.1.3 Liquidity Risk 11
2.1.4 Operational Risk 12
2.1.5 Legal Risk 12
2.2 BASLE Accords version 1 and 2 13
2.2.1 Basle Accord 1988 (version 1) 13
2.2.2 Basle 1996 Amendment (version II) 15
2.2.3 Taiwan Regulated Capital Requirements 17
2.3 VaR & Methodologies 18
2.3.1 GARCH 19
2.3.2 The RiskMetricsTM Approach 21
2.3.3 Historical Simulation Approach 22
2.3.4 Monte Carlo Simulation Approach 23
Chapter 3 Data 25
3.1 Formalization of Portfolio A and B — Process and Assumptions 25
3.1.1 Assumptions for Sample Size & Allocation of Asset Instrument 25
3.1.2 Assumptions for Portfolio A & B — Holding Period 28
3.2 Daily P & L (Mark-to-Market Results) 29
Chapter 4 Methodologies 30
4.1 GARCH (1,1) Model 30
4.2 Testing Model Performance 33
4.3 Unconditional and Conditional Coverage Models 35
Chapter 5 Empirical Results 37
5.1 Daily P & L Time Series Return 37
5.2 Performance Testing Results Under Various VaR Approach 37
5.3 Model Efficiency vs. Capital Charges 38
Chapter 6 Conclusions 40
References 43
Appendix I The standard structure and formula for RisMetricsTM, Historical Simulation and Monte Carlo Simulation …. Sourced from Taiwan Economic Journal White Papers 70

1. Basle Committee on Banking Supervision, 1996b, “Amendment to the capital accord to incorporate market risks. Bank for International Settlements”
2. Basle Committee on Banking Supervision, 1996a, “Supervisory Framework for the Use of Backtesting in Conjunction with the internal Models Approach to Marke Risk Capital Requirements.” Report#22 , Bank for International Settlements.
3. Berkowitz J & O’Brien J ,2002, “How Accurate Are Value-at-Risk” Models at Commercial Banks?” , The Journal of Finance Vol. LVII, No. 3 June 2002
4. Basak S & Sharpiro A 2001, “Value-at-Risk-based Risk Management: Optimal Policies and Asset Prices”, The Review of Financial Studies Summer 2001, Vol 14
5. Bams Dennis 2001, “An Evaluation Framework for Alternative VaR Models”
6. Berkowitz, Jeremy. 2000. “A Coherent Framework for Stress Testing,:, Journal of Risk 2 (Winter), 5-15
7. Berkowitz,Jeremy,2001, “Testing density forecasts with applications to risk management”, Journal of Business and Economic Statistics 19, 465-474.
8. Berkowitz, J. (2001), “ Testing the Accuracy of Density Forecasts, Journal of Business and Economic Statistics.
9. Board of Governors of the Federal Reserve System, 1995, Request for Comment on the Pre-Commitment Approach for Market Risks, Board of Governors of the Federal Reserve System, Docket No. R-0886, Washington, D.C.
10. Boyle,P., 1977 “Options : A Monte Carlo Approach, “ Journal of Financial Economics 4, 323-338
11. Butler, J.S., and Barry Schachter, 1998, “Estimating Value at Risk with a Precision Measure by Combining Kernel Estimation with Historical Simulation,” Review of Derivatives Research 1, 371-390
12. Bollerslev t., 1986, “Generalized Autoregressive Conditional Heteroskedasticity,” Jounal of Econometrics 31, 307-327
13. Bollerslev.T., Engle,R.F. and Wooldridge, J.M. 1988. “ A Capital Asset Pricing Model with Time-Varying Covariances,” Journal of Political Economy, 96, 116-131
14. Bollerslev.T.,R Chou, and K. Kroner, 1992, “ARCH Modelling in Finance: A Review of Theory and Empirical Evidence, “ Journal of Econometrics 52, 5-59
15. Bollerslev.T.,and Ghysels, E., 1996 “Periodic Autoregressive Conditional Heteroskedasticity ,” Journal of Business and Economic Statistics, 14, 139-157.
16. Christoffersen ,P. (1986), “Evaluating Interval Forecasts,” International Economic Review, 29, 841-862
17. Christoffersen, P. and F.X. Diebold (2000), “How Relevant is Volatility Forecasting for Financial Risk Management,” Review of Economics and Statistics , 82, 12-22
18. Christoffersen P, Hahn J and Inoue A, 1999, “Testing, Comparing, and Combining Value-at-Risk Measures”, Oct. 1, 1999
19. C.O Alexander and C.T. Leigh. On the covariance matrices used in value at risk models. The Journal of Derivatives, page 50-62, Sprint 1997
20. Christiansen Charlotte 1998, “Value at Risk Using the Factor-Arch Model” URL www.hha.dk/cha/
21. Central Bank of China web site http://cbc.gov.tw
22. Diebold, F.X., Gunther, T.A. and Tay, A.S. (1998) , “ Evaluating Density Forecasts,” International Economic Review, 39, 863-883
23. Engle, R.F. and S. Manganelli (1999), “CAViaR : Conditional Autoregressive Value-at-Risk by Regression Quantiles,” manuscript, University of California, San Diego.
24. Giovanni Barone-Adesi,Kostas Giannopoulos, Les Vosper, 1999, “VaR without Correlations for Portfolios of Derivative Securities”, The Journal of Futures Markets, Vol. 19, No. 5, 583-602 (1999)
25. Jorion P, “Value at Risk-The new Benchmark for Managing Financial Risk”, Second Edition., published by McGraw-Hill
26. Kupiec, P. (1995). “Techniques for Verifying the Accuracy of Risk Measurement Models,” Journal of Derivatives, 3, 73-84
27. Kupiec,Paul, 1998, “Stress Testing in a Value at Risk Framework,” Journal of Derivatives 6 (Fall), 7-24
28. Lopez, Jose , 1999, “Regulatory Evaluation of Value-at-Risk Models,” Journal of Risk , 1 (Winter), 37-63
29. Lopez ,J. and Walter, C. (2000) “Evaluating Covariance Matrix Forecast in a Value-at-Risk Framework,” manuscript, Federal Reserve Bank of San Francisco.
30. McNeil, Alexander, 1999, “Extreme Value Theory for Risk Managers,” in internal Modeling and CADII, Risk Publications , London , pp93-113
31. Matten, Chgris, 1996, “Managing Bank Capital : Capital Allocation and Performance Measurement, “ Wiley, Chichester, England.
32. Pritsker,Mattew,1997, Evaluating Value-at-Risk methodologies : Accuracy versus computational time, Journal of Financial Services Research 12, 201-242
33. Picoult, Evan, 1997, “Calculating Value at Risk with Monte Carlo Simulation, “ in Risk Management for Financial Institutions , Risk Publications, London pp. 73-92
34. Risk MetricsTM , third edition, J.P. Morgan 1995
35. Shirrell, David, 1998, “The Rise and Rise of the Risk Manager,” Euromoney (February) , 56-60
36. Taiwan Minister of Finance web site http://www.boma.gove.tw
37. Taiwan Economic Journal web site
38. Wilson, Duncan, 1995, “VAR in Operation,” Risk 8 (December) , 24-25
39. Zangari, P., 1997, “Streaming the Market Risk Measurement Process,” RiskMetrics Monitor, 1, 29-35