共移效果、多重結構改變及預測— 波動指數相關研究

本論文在於探討波動指數及其衍生的金融商品對於標的商品的影響。運用各國波動指數(VIX)、指數股票型基金的波動指數(ETF-VIX)、波動指數的指數股票型基金(VIX-ETF)以及個股權益波動指數(Equity-VIX)進行實證研究，冀望透過波動指數與股票指數(S&P 500)的關係，能夠衍伸運用於其他相關的波動指數，進而達到預測金融商品的趨勢，以利金融風險的控管。第一篇論文探討重心在於各國權益指數與波動指數的共移效果，運用全球向量自我回歸模型(GVAR)，結果顯示波動指數具有領先於權益指數的效果。第二篇論文著重指數股票型基金的波動指數，以平滑轉換自我回歸模型驗證，是否能提供指數股票型基金 (ETF)的預測能力，結果顯示指數股票型基金的波動指數與指數股票型基金的關係不如權益指數與波動指數的關係。第三篇論文對追蹤波動指數的指數股票型基金之緩長記憶和多重結構性改變之檢測，實證結果顯示發生多重結構改變但是並沒有產生雙重緩長記憶模型。第四篇論文是納入相關變數運用灰關聯及類神經模型，來預測個股權益波動指數，結果顯示情緒指數對於他的影響最重要，而個股股價最不重要；另外以回饋式類神經網路(RNN)模型預測最精準。

關鍵字

波動指數、全球向量自我回歸模型、指數股票型基金的波動指數、平滑轉換自我回歸模型、多重結構改變、緩長記憶、股權益波動指數、類神經模型

並列摘要

This thesis examined the volatility index (VIX) and derivative volatility indexes of financial commodities impact on underlying assets, applying VIX of signal country, commodities ETF-VIX, VIX-ETF and the equity-VIX. Through the relationship between volatility index and S&P 500 could be able to extend for other related volatility indexes, thus achieving the predicted trend of financial products in order to control risks. The first essay researched the international economic linkages among the US and other countries using Global VAR (GVAR) model with individual country’s volatility index and stock index. The findings revealed that the volatility index had the leading effect to equity indices on international economic co-movement. The second essay examined the relationship between ETF-VIX and ETF applying STRA(X) model to forecast ETF. An important conclusion of the study is that commodity ETF-VIX and ETF had very weak correlation, unlike that in the stock market. The third essay illustrated whether VIX-ETF with multiple structural changes and efficient market hypothesis utilizing Bai-Perron、Iterated Cumulative Sums of Squares (ICSS) and ARFIMA-FIGARCH models. The results were not all in line with market efficiency, especially in the dual long memory model. The fourth essay focused on the effect of predication on the individual equity volatility index process exercising Gray Relational Analysis (GRA) and Artificial Neural Networks (ANNs) models. The results showed that the most important factor was sentiment index, but the less important factor was individual stock price; at the same time the Recurrent Neural Network (RNN) model generated the most accurate predicting performance.

並列關鍵字

ANN ； VIX ； Global VAR ； ETF-VIX ； STRA(X) ； Long Memory ； Equity-VIX

參考文獻

Dong, J. and Z. Zhang (2005). "Chinese GDP and Financial Analyzing via Grey Relational Analysis." Journal of Grey System 17(4): 383-386.

Wiphatthanananthakul, C. C. and S. Sriboonchitta (2010). "ARFIMA-FIGARCH and ARFIMA-FIAPARCH on Thailand Volatility Index." International Review of Applied Financial Issues and Economics 2(1): 193-212.

Kumar, R. and H. Gupta (2009). "Volatility in the Indian Stock Market: A Case of Individual Securities." Journal of Academic Research in Economics 1(1): 34-45.

Mukul, Mukesh Kumar, Vikrant Kumar and Sougata Ray (2012). "Gold ETF Performance: A Comparative Analysis of Monthly Returns." IUP Journal of Financial Risk Management 9(2): 59-63.

Engle, R. F. and D. Sarkar (2002). "Pricing Exchange Traded Funds." NYU Working Paper No. S-DRP-02-11.

國際替代計量

共移效果、多重結構改變及預測— 波動指數相關研究

全文下載

主題瀏覽