Evaluation of Value-At-Risk in Investment Portfolio

• Main Authors: Liu, Ming Yi, 劉銘益
• Other Authors: 廖四郎
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/54082816664738866277

碩士 === 國立政治大學 === 統計學系 === 104 === Recently, with the development of liberalization and globalization in financial markets, investor is faced with more investment opportunity and investment risk simultaneously, and this makes investor evaluate VaR. Therefore, it becomes the most significant topic for investor to utilize the concept of investment portfolio to select and adopt suitable risk measure method to evaluate risk and further control risk. Based on this reason, this study combines the concept of VaR with the theory of portfolio. This thesis utilize Markowitz’s Mean-Variance approach and Growth Value Index(GVI) to select each optimal stock portfolio from Taiwan’s Financial Stock、Taiwan’s traditional Stock、Taiwan’s Electronic Stock and publicly traded company in Taiwan Stock Exchange. Furthermore, employing Historical Simulation Approach、Variance-Covariance Approach and GARCH to evaluate the VaR of that optimal portfolio. Finally, through Kupiec test (1995) to evaluate each model’s forecasting performance. Empirical study shows that from the results of the out-of-sample forecasting, we can find that GARCH is the best one to forecast the VaR, and decompose the portfolio VaR with marginal VaR and component VaR .This kind of analysis can provide managers with more accurate decision in making risk management.