DOI QR코드

DOI QR Code

Validity assessment of VaR with Laplacian distribution

라플라스 분포 기반의 VaR 측정 방법의 적정성 평가

  • Byun, Bu-Guen (Department of Electronic, Information, and Communication Engineering, Hongik University) ;
  • Yoo, Do-Sik (Department of Electronic, Information, and Communication Engineering, Hongik University) ;
  • Lim, Jongtae (Department of Electrical, Information, and Control Engineering, Hongik University)
  • 변부근 (홍익대학교 전자정보통신공학과) ;
  • 유도식 (홍익대학교 전자정보통신공학과) ;
  • 임종태 (홍익대학교 전기정보제어공학과)
  • Received : 2013.08.27
  • Accepted : 2013.10.01
  • Published : 2013.11.30

Abstract

VaR (value at risk), which represents the expectation of the worst loss that may occur over a period of time within a given level of confidence, is currently used by various financial institutions for the purpose of risk management. In the majority of previous studies, the probability of return has been modeled with normal distribution. Recently Chen et al. (2010) measured VaR with asymmetric Laplacian distribution. However, it is difficult to estimate the mode, the skewness, and the degree of variance that determine the shape of an asymmetric Laplacian distribution with limited data in the real-world market. In this paper, we show that the VaR estimated with (symmetric) Laplacian distribution model provides more accuracy than those with normal distribution model or asymmetric Laplacian distribution model with real world stock market data and with various statistical measures.

VaR (value at risk)는 주어진 신뢰수준에서 일정기간 동안 발생할 수 있는 최대손실의 기대치를 나타내는 것으로, 현재 금융기관들의 대표적인 위험관리 수단으로 사용되고 있다. 기존의 대다수 연구에서는 수익률의 확률분포를 정규분포라 모형화한 후 VaR을 측정한다. 최근 Chen 등 (2012)은 수익률의 확률분포를 비대칭 라플라스 분포라 모형화하고 VaR을 측정하였기도 하였으나, 비대칭 라플라스 분포의 경우 그 모양을 결정하는 최빈값, 비대칭 정도, 분산정도 등을 실제적인 환경에서 제한된 개수의 데이터를 가지고 추정하기가 매우 어렵다는 단점이 있다. 이 논문에서, 우리는 (대칭) 라플라스 분포 모형이 정규분포 모형이나 비대칭 라플라스 분포 모형보다 실제적인 환경에서 VaR을 보다 더 정확히 추정해 줌을 주식시장의 실제 데이터와 VaR 초과비율, 기대초과손실, VaR초과편차율 등의 통계지표를 도입하여 입증한다.

Keywords

References

  1. Chen, Q, Gerlach, R. and Lu, Z. (2012). Bayesian value-at-risk and expected shortfall forecasting via the asymmetric Laplacian distribution. Computational Statistics and Data Analysis, 56, 3498-3516. https://doi.org/10.1016/j.csda.2010.06.018
  2. Cho, D. (2004). The effects of estimation methods of stock price volatility on VaR. Korean Journal of Futures and Options, 12, 1-24.
  3. Kim, E. and Lee, T. (2011). A numerical study on portfolio VaR forecasting based on conditional copula. Journal of the Korean Data & Information Science Society, 22, 1065-1074.
  4. Jorion, P. (1997). Value at risk: The new benchmark for controlling market risk, 1st Ed., McGraw-Hall, New York.
  5. J. P. Morgan and Reuters (1996). Riskmetrics-Technical document, 4th Ed., Morgan Guaranty Trust Company, New York.
  6. Lee, J. (2000). Backtesting of VaR estimates and evaluation for VaR model. Korean Journal of Futures & Options, 8, 81-106.
  7. Lee, K. (2006). A comparison of the out-of-sample performance of VaR models. Korean Journal of Money & Finance, 11, 127-168.
  8. Lu, Z., Huang H. and Gerlach, R. (2010). Estimating value at risk: From JP Morgan's standard-EWMA to skewed-EWMA forecasting, Working Paper, Faculty of Economics and Business, The University of Sydney.
  9. Ohk, K., Huh, H. and Cheon, S. (2012). A comparison of value at risk by the estimation models of KOSPI200 return volatility. Journal of Finance, 5, 21-39.
  10. Yoo, J. and Lee, J. (2002). Price limits and VaR estimation of the KOSDAQ 50 companies. Asia-Pacific Journal of Financial Studies, 30, 73-99.

Cited by

  1. On multivariate GARCH model selection based on risk management vol.25, pp.6, 2014, https://doi.org/10.7465/jkdi.2014.25.6.1333
  2. Saddlepoint approximations for the risk measures of portfolios based on skew-normal risk factors vol.25, pp.6, 2014, https://doi.org/10.7465/jkdi.2014.25.6.1171
  3. Optimal portfolio and VaR of KOSPI200 using One-factor model vol.26, pp.2, 2015, https://doi.org/10.7465/jkdi.2015.26.2.323