應用已實現波動度於選擇權評價的實證研究

Abstract

先前文獻已提出：利用高頻率資料的異質性自我相關已實現波動度(HAR-RV)模型，較其他波動度模型更能捕捉財務市場報酬波動度的特性及更準確的預測波動度。然而，就本文所知，直到目前為止尚未有研究探討應用HAR-RV模型於選擇權中，是否可減少選擇權評價誤差及增進選擇權delta動態避險績效。此外，過去的實證結果發現EGARCH模型在選擇權評價上優於其他波動度模型。 因此，本文研究目的為：將HAR-RV與EGARCH選擇權評價模型用於S&P500指數選擇權的評價，並比較二模型在評價及避險績效上的差異。本文實證結果發現：第一、除了樣本外的價外買權和價外賣權，HAR-RV模型在樣本外的買權和賣權的評價誤差較EGARCH模型小；第二、HAR-RV模型在樣本外的買權避險績效較佳，而EGARCH模型在樣本外的賣權避險績效表現較好，然而，此樣本外的賣權避險績效並沒有顯著地比HAR-RV模型的樣本外賣權避險績效佳。

Previous studies have documented that, with use of high frequency data, Heterogeneous Autoregressive of the Realized Volatility (HAR-RV) model performs better than other volatility models in fitting financial return volatility measurement and has a more accurate forecasting ability. However, to our knowledge, no previous studies have investigated whether the HAR-RV model can improve option pricing and delta dynamic hedging performance in financial markets. Additionally, previous empirical analysis of option pricing models with the framework of EGARCH have presented superior to other volatility models. Using S&P 500 index options data, this study compares the HAR-RV and the EGARCH option pricing model in terms of option pricing and dynamic hedging performance. As expected, the results of this study demonstrate that the HAR-RV option pricing model is superior in terms of out-of-sample call and out-of-sample put option pricing performance for all moneyness except for out-of-the-money options. In out-of-sample hedging performance, the HAR-RV model still performs better than the EGARCH model, except in the case of put options. However, the EGARCH option pricing model does not show significant superiority in hedging performance of put options.