This study develops a novel approach for improving stock return volatility forecasts using volatility index information with the entropic tilting technique. Unlike traditional linear heteroskedasticity autoregressive methods with option-implied information, we first derive predictive densities from traditional models, and then tilt using both the first and second moments of the risk-neutral distribution, which enables us to capture the nonlinear effect in our specification. The empirical findings demonstrate a substantial enhancement in the forecasting accuracy of all models once the first- and second-moment information is considered, where the improvement is both statistically and economically significant. These results have important implications for risk management in well-established derivatives markets.

This paper introduces a novel hedging strategy based on textual information for cross-hedging. The strategy allows the optimal hedge ratio to vary with the sentiment extracted from online oil news, leading to an improvement in cross-hedging effectiveness. We find that the sentiment hedging strategy outperforms traditional time-varying and static hedging strategies in both in-sample and out-of-sample scenarios, with statistical and economic significance. This superior performance benefits from the well-documented predictive power of sentiment on futures volatility. Furthermore, we confirm an asymmetric effect of sentiment on hedging effectiveness, with negative sentiment having a greater impact on hedge ratios. Finally, extreme sentiment can trade off the negative relationship between volatile hedge ratios and hedging performance, potentially leading to superior hedging performance compared to normal sentiment, which is robust when considering transaction costs. Our findings highlight the importance of sentiment for estimating the optimal hedge ratio and should be considered for risk management.

This paper aims to explore the extent to which text data contains valuable information for predicting oil futures returns. A novel mixed-frequency data sampling random forest regression (MIDAS-RF) approach is proposed to construct a textual indicator. This approach can extract nonlinearity and interaction information from news and allows us to better handle the mixed-frequency and high-dimensional data. Comparing it with traditional sentiment variables and financial factors, our indicator demonstrates better forecasting performance both statistically and economically, with a monthly out-of-sample R2 of 5.26% and an annualized certainty equivalent return gain of 3.08%, respectively. Further evidence suggests that the predictability of the textual indicator is primarily driven by words related to capital markets and macroeconomic topics.

We develop a new approach that shrinks a given model forecast to the benchmark model forecast in order to improve forecasting performance. Simulation results show the superior performance of our approach, relative to popular methods such as forecast combination and the robustness to model misspecification. We apply our method to forecasting the returns on the S&P 500 index and find significant predictability when shrinking the principal component (PC) regression forecasts based on statistical and economic evaluation criteria. The forecast improvement from our shrinkage approach can be explained by the ability of its hyperparameters to be better predict real economic changes.

In this paper, we propose an affine discrete-time model that incorporates the jump process and spillover effect for valuing the 50 ETF options in China. Based on the proposed model, a closed-form solution is also derived for the new dynamics of underlying asset, which facilitates option pricing. The empirical results show that the proposed model offers greater economic benefit with reduced pricing errors than the traditional benchmark models, including the popular HNGARCH model of Heston and Nandi (2000), GARV model of Christoffersen et al. (2014), and BPJVM model of Christoffersen et al. (2015). Our finding is important for financial risk management and investment in Chinese derivatives market.

While several theoretical models imply that uncertainty has predictive ability for stock returns, few studies investigate this issue using empirical data. We fill this gap by comparing the predictive ability of uncertainty variables with the predictive ability of well-known economic level variables. We find the in-sample and out-of-sample return predictability using the combining uncertainty information. The predictability is significant from both economic and statistical perspectives. Further analysis shows that macroeconomic uncertainty and level information provide complementary predictive ability over the business cycle. We obtain stronger and more robust return predictability using both types of information together than using either source of information alone.

Macroeconomic uncertainty variables have significant effects on long term components of stock price volatility and can be used to improve VaR and ES prediction for a given stock

In this paper, we develop a new volatility model capturing the effects of macroeconomic variables and jump dynamics on the stock volatility. The proposed GARCH-Jump-MIDAS model is applied to the S&P 500 index. Our in-sample results indicate that macroeconomic activities have important impacts on aggregate market volatility. Out-of-sample evidence suggests that our model with macroeconomic variables significantly outperform a wide range of competitors including the original GARCH(1,1), GARCH-MIDAS and GJR-A-MIDAS models. The volatility timing results also show that the information from jumps and macroeconomic activity is helpful for improving the portfolio performance.

We develop a novel method to impose constraints on univariate predictive regressions of stock returns. Unlike previous approaches in the literature, we implement our constraints directly on the predictor, setting it to zero whenever its value falls within the variable's past 24-month high and low. Empirically, we find that relative to standard unconstrained predictive regressions, our approach leads to significantly larger forecast gains. We also show how a simple equal-weighted combination of our constrained forecasts leads to further improvements in forecast accuracy, generating forecasts that are more accurate than those obtained using current constrained methods. Further analysis confirms that these findings are robust to the presence of model instabilities and structural breaks.

We develop a new generalized autoregressive conditional heteroskedasticity (GARCH) model that accounts for the information spillover between two markets. This model is used to detect the usefulness of the CBOE volatility index (VIX) for improving the performance of volatility forecasting and option pricing. We find the significant ability of VIX to predict stock volatility both in‐sample and out‐of‐sample. VIX information also helps to greatly reduce the option pricing error. The proposed volatility spillover GARCH model performs better than the related approaches proposed by Kanniainen et al. (2014, J Bank Finance, 43, pp. 200‐211) and P. Christoffersen et al. (2014, J Financ Quant Anal, 49, pp. 663–697).

随着我国金融市场逐步开放,国内外机构投资者数量与规模呈递增趋势.本文基于金融自由化视角,考察国内外机构投资者市场交易与股票市场定价效率的关系,研究结果显示,国外机构投资者能够促进股票市场定价效率,国内机构投资者对股票市场定价效率有一定的异质性特征,国外机构投资者可以通过产生定价效率技术溢出效应提高国内机构投资者对股票市场定价效率.

This paper proposes a new measure of belief dispersion for the Chinese stock market based on the closing price data from mobile and PC trading terminals. Our results show that our belief dispersion measure has significant predictive content for aggregate market volatility both in-sample and out-of-sample. The volatility predictability is robust to different realized volatility measures, forecasting horizons, and benchmark models. Our belief dispersion measure also helps improve the density prediction. Overall, we find that investors with mean-variance preferences who use belief dispersion information to generate volatility forecasts can improve their portfolio performance over longer horizons.

Option prices contain important information about risk preferences. This study proposes an option-based model to estimate the optimal dynamic hedging ratio. Using a sample of S&P 500 index, we find that the option-implied hedging ratio has the best performance both in-sample and out-of-sample due to its relative risk aversion. This finding will help risk managers reduce their hedging risk.

Taking account of some stylized facts in correlation processes and feasible implementation, a new regime switching dynamic equicorrelation (RS-DEC) model is proposed. Estimate procedures and large sample properties are also provided. RS-DEC model not only deals with high-dimensional correlation, but also takes account for structure break and asymmetry in correlation. In empirical work, we examine the asset allocation on 97 stocks in the Shanghai Stock Exchange. Our model can provide a better fit in sample, and give the information for correlations structure break; Comparing with Na(i)ve strategy, under Sharpe ratio and minimum standard error criteria, the results show that our model can improve the out-of-sample performance, and significant tests support these conclusions.

In this paper, we extend the GARCH-MIDAS model proposed by Engle et al. (2013) to account for the leverage effect in short-term and long-term volatility components. Our in-sample evidence suggests that both short-term and long-term negative returns can cause higher future volatility than positive returns. Out-of-sample results show that the predictive ability of GARCH-MIDAS is significantly improved after taking the leverage effect into account. The leverage effect for short-term volatility component plays more important role than the leverage effect for long-term volatility component in affecting out-of-sample forecasting performance.

Marginal distribution requires to be tested in term of Copula theory,but the test can be improved in empirical analysis,which reduces the risks of model.Unlike testing apart,we use a new simultaneous statistic for testing the marginal distribution,Monte Carlo results show that our method is more robust than testing apart.From asset allocation analysis,we find that the marginal distribution which passes the test can obtain higher efficient frontier,and get a better out-of-sample performance.The significant test supports our conclusions.

In this paper, we introduce the functional coefficient to heterogeneous autoregressive realized volatility (HAR‐RV) models to make the parameters change over time. A nonparametric statistic is developed to perform a specification test. The simulation results show that our test displays reliable size and good power. Using the proposed test, we find a significant time variation property of coefficients to the HAR‐RV models. Time‐varying parameter (TVP) models can significantly outperform their constant‐coefficient counterparts for longer forecasting horizons. The predictive ability of TVP models can be improved by accounting for VIX information. Copyright © 2016 John Wiley & Sons, Ltd.

2003年中国开始引入境外合格的证券投资机构(QFII),其对于我国建立有效率的资本市场具有重要意义.本文基于2006-2015申万一级行业的数据,运用面板数据回归方法考察QFII对A股定价效率的影响,发现QFII投资能够提高中国股票的定价效率,但是QFII投资规模与股票定价效率之间不是单调的;随着QFII投资额度超过临界值,本国证券市场定价效率反而会下降;此外,在不同的行情中QFII投资中国证券市场,其定价效率有一定的异质性.因此,为了有效地避免和化解金融风险,QFII 的投资规模应该控制在一定的临界值之下.

We introduce a regime switching GARCH-MIDAS model to investigate the relationships between oil price volatility and its macroeconomic fundamentals. Our model takes into account both effects of long-term macroeconomic factors and short-term structural breaks on oil volatility. The in-sample and out-of-sample results show that macroeconomic fundamentals can provide useful information regarding future oil volatility beyond the historical volatility. We also find the evidence that the structural breaks cause higher degree of GARCH-implied volatility persistence. Two-regime GARCH-MIDAS models can significantly beat their single-regime counterparts in forecasting oil volatility out-of-sample.

Predictability of macroeconomic and financial variables is an important issue in economics. In this paper, we propose a nonparametric test for the predictability of the direction of price changes. The Monte Carlo simulation results show that our method displays better finite-sample property than the traditional parametric Granger causality test (Granger, 1969) and two nonparametric causality tests of Hiemstra and Jones (1994) and Diks and Panchenko (2006). (C) 2016 Elsevier B.V. All rights reserved.