財務市場波動及基金績效之計量分析

Abstract

本研究主要利用計量分析探討兩個財務市場的重要議題。第一個議題是財務市場波動，利用 HAR 及 MIDAS 迴歸模型探討已實現變幅波動的預測績效，研究結果發現已實現變幅波動較已實現波動有效，將波動分為連續與跳躍作為迴歸子預測未來的波動幾乎較利用其他不同變異的迴歸子其MSE為最小，並檢定出在已實現變幅波動下短期的跳躍會產生結構性的改變。此外，並以 HAR 及 MIDAS 迴歸模型為涵蓋迴歸，將已實現變幅波動分成連續與跳躍部份，探討此二者及隱含波動對已實現變幅波動的資訊內涵，研究結果顯示隱含波動對已實現變幅波動具有很高的資訊內涵，連續部份提供與隱含波動部份相同的資訊內涵，然而跳躍部份無法對已實現變幅波動提供任何有用的資訊；若僅考慮隱含波動與樣本外預測，則隱含波動亦提供較大的資訊內涵。第二個議題是利用門檻迴歸模型檢驗基金經理人的選股及擇時能力，實證結果顯示傳統的Henriksson 及 Merton (1981) 模型為我們所提的門檻迴歸模型之特例，利用傳統的迴歸模型會低估擇時能力，而門檻迴歸模型能產生較為正確的推論。

This dissertation consists of two separate issues. The first issue is to discuss the forecasting performance of HAR and MIDAS regression models of realized range-based volatility; we focus on the S&P 500 index. The empirical results show that the realized range-based volatility is more efficient than the realized return-based volatility; the regressors consisting of the continuous sample path and jump variability measures in the HAR and MIDAS regressions predict the future realized range volatilities, and thus dominate almost in all MSE terms. In addition, the realized range-based regressions are significant for short-run volatility forecasting, but the realized return-based regressions are almost invariant to jumps. Furthermore, we will employ the HAR and MIDAS regressions as encompassing regressions to examine the information content of the continuous and jump components of the realized range-based volatility, and the additional information content of the implied volatility as an additional regressor. We use the VIX as the measure of the implied volatility. We find that the implied volatility has a high information content and the past continuous components feature relevant information content by the implied volatility. Besides, the jump components do not contribute to future valuable information. The second issue is to detect mutual fund market timing abilities, using the threshold regression model. The empirical results show that the traditional Henriksson and Merton (1981) model is only a special case within our model, and we demonstrate the potential bias of using the traditional model, arguing that it tends to underestimate the market-timing effect. Indeed, we find that the use of the traditional market timing test may provide misleading results in some circumstances; thus, our proposed threshold model provides more accurate inferences on the market-timing effects of mutual funds.