主成份分析方法的最新發展與應用

The Latest Development and Application of Principal Component Analysis

陳泓宇(Hung-Yu Chen)

指導教授 ： 林建甫

國立臺灣大學/社會科學院/經濟學研究所/碩士(2021年)

https://doi.org/10.6342/NTU202104429

摘要

當投資者彼此決策相關性很高時，在事件的驅動下有較高機率觸發羊群效應使投資者同步賣出引發系統性風險。本文使用主成分分析產出不同參數天期的第一主成分解釋能力和第一主成分解釋能力變化，以此代表證券市場的相關性，並以兩項指標的上升預測自2005年來下跌超過20%之系統性風險事件，研究發現主成分分析產出之指標可以預測系統性風險，在和台指選擇權波動率指標相比後也有較佳的預測能力。

關鍵字

主成分分析 ； 系統性風險

並列摘要

When investors in the financial market have highly correlated descision, there is a higher probability to trigger systemic risks by herd effect. This study uses principal component analysis to produce the first principal component explains and change of the first principal component explains, and uses the two indicators to predict systemic risk events that have fallen by more than 20% since 2005. The study found that indicators can predict systemic risks, and both indicators have better predictive ability compared with the TAIEX Options Volatility Index.

並列關鍵字

PCA ； Systemic risk

參考文獻

邱永金與郭玟秀(2008), "台指選擇權隱含波動率指標對真實波動率與指數報酬的資訊內涵研究", 嶺東科技大學碩士論文

駱怡帆 (2008) , "市場情緒是否影響證券報酬？東亞新興市場證券波動指數間之計量因果性", 銘傳大學碩士論文

Avellaneda, Marco, and Lee, Jeong Hyun (2010), "Statistical Arbitrage in the US Equities Market," Quantitative Finance, Taylor Francis Journals, 10(7), 761-782.

Avellaneda, Marco, and Serur, Juan Andrés (2020), "Hierarchical PCA and Modeling Asset Correlations," Papers 2010.04140, arXiv.org.

Butler, K.C., and Joaquin, D.C., (2002), "Are the Gains from International Portfolio Diversification Exaggerated? The Influence of Downside Risk in bear markets," Journal of International Money and Finance, Elsevier, 21(7), 981-1011