Return distributions and volatility forecasting in metal futures markets: Evidence from gold,silver, and copper

The characterization of return distributions and forecast of asset‐price variability play a critical role in the study of financial markets. This study estimates four measures of integrated volatility—daily absolute returns, realized volatility, realized bipower volatility, and integrated volatility via Fourier transformation (IVFT)—for gold, silver, and copper by using high‐frequency data for the period 1999 through 2008. The distributional properties are investigated by applying recently developed jump detection procedures and by constructing financial‐time return series. The predictive ability of a GARCH (1,1) forecasting model that uses various volatility measures is also examined. Three important findings are reported. First, the magnitude of the IVFT volatility estimate is the greatest among the four volatility measures. Second, the return distributions of the three markets are not normal. However, when returns are standardized by IVFT and realized volatility, the corresponding return distributions bear closer resemblance to a normal distribution. Notably, the application of financial‐time sampling technique is helpful in obtaining a normal distribution. Finally, the IVFT and realized volatility proxies produce the smallest forecasting errors, and increasing the time frequency of estimating integrated volatility does not necessarily improve forecast accuracy. © 2010 Wiley Periodicals, Inc. Jrl Fut Mark 31:55–80, 2011     

中国股市波动的CARR模型分析

ARCH/GARCH模型在波动性的预测已被学者广泛使用并在实证上得到良好的效果。本文以上海股市为研究对象,分别运用CARR模型和GARCH模型进行波动性预测,进而对两种方法的预测能力进行比较,实证结果表明CARR模型在拟合波动性方面优于GARCH模型。     

Stock return dynamics,option volume,and the information content of implied volatility

This article reports new empirical results on the information content of implied volatility, with respect to modeling and forecasting the volatility of individual firm returns. The 50 firms with the highest option volume on the Chicago Board Options Exchange between 1988 and 1995 are examined. First, the results indicate that the ability of implied volatility to subsume all relevant information about conditional variance depends on option trading volume. For the most active options in the sample, implied volatility reliably outperforms GARCH and subsumes all information in return shocks beyond the first lag. For these active options, implied volatility performs substantially better than indicated by the prior results of Lamoureux and Lastrapes ( 1993 ), despite significant methodological improvements in the time‐series volatility models in this study including the use of high‐frequency intraday return shocks. For the lower option‐volume firms in the sample, the performance of implied volatility deteriorates relative to time‐series volatility models. Finally, compared to a time‐series approach, the implied volatility of equity index options provides reliable incremental information about future firm‐level volatility. © 2003 Wiley Periodicals, Inc. Jrl Fut Mark 23:615–646, 2003     

VIX futures and its closed‐form pricing through an affine GARCH model with realized variance

This paper studies the forecasting of volatility index (VIX) and the pricing of its futures by a generalized affine realized volatility model proposed by Christoffersen et al. This model is a weighted average of a GARCH and a pure realized variance (RV) model that incorporates each volatility component into the new dynamics. We rewrite the VIX in terms of both volatility components and then derive closed‐form formulas for the VIX forecasting and its futures pricing. Our empirical studies find that a unification of the GARCH and the RV in the modeling substantially improves the forecasting of this index and the pricing of its futures.     

Measuring implied volatility: Is an average better? Which average?

Options researchers have argued that by averaging together implied standard deviations, or ISDs, calculated from several options with the same expiry but different strikes, the noise in individual ISDs can be reduced, yielding a better measure of the market's volatility expectation. Various options researchers have suggested different weighting schemes for calculating these averages. In the forecasting literature, econometricians have made the same argument but suggested quite different weighting schemes. Ignoring both literatures, commercial vendors calculate ISD averages using their own weightings. We compare the averages proposed in both the options and econometrics literatures and the averages used by major commercial vendors for the S&P 500 futures options market. Although some averages forecast better than others, we find that the question of the best weighting scheme is of secondary importance. More important is the fact that the ISDs are upward biased measures of expected volatility. Fortunately, this bias is stable over time, so past bias patterns can be used to obtain unbiased volatility forecasts. Once this is done, most ISD averages forecast better than time series and naive models, and the differences between the averages produced by the various proposed weighting schemes are small. © 2002 Wiley Publications, Inc. Jrl Fut Mark 22:811–837, 2002     

Value at risk and conditional extreme value theory via markov regime switching models

This study develops a new conditional extreme value theory‐based (EVT) model that incorporates the Markov regime switching process to forecast extreme risks in the stock markets. The study combines the Markov switching ARCH (SWARCH) model (which uses different sets of parameters for various states to cope with the structural changes for measuring the time‐varying volatility of the return distribution) with the EVT to model the tail distribution of the SWARCH processed residuals. The model is compared with unconditional EVT and conditional EVT‐GARCH models to estimate the extreme losses in three leading stock indices: S&P 500 Index, Hang Seng Index and Hang Seng China Enterprise Index. The study found that the EVT‐SWARCH model outperformed both the GARCH and SWARCH models in capturing the non‐normality and in providing accurate value‐at‐risk forecasts in the in‐sample and out‐sample tests. The EVTSWARCH model, which exhibits the features of measuring the volatility of a heteroscedastic financial return series and coping with the non‐normality owing to structural changes, can be an alternative measure of the tail risk. © 2008 Wiley Periodicals, Inc. Jrl Fut Mark 28:155–181, 2008     

Nonlinear asymmetric models of the short‐term interest rate

This study introduces a generalized discrete time framework to evaluate the empirical performance of a wide variety of well‐known models in capturing the dynamic behavior of short‐term interest rates. A new class of models that displays nonlinearity and asymmetry in the drift, and incorporates the level effect and stochastic volatility in the diffusion function is introduced in discrete time and tested against the popular diffusion, GARCH, and level‐GARCH models. Based on the statistical test results, the existing models are strongly rejected in favor of the newly proposed models because of the nonlinear asymmetric drift of the short rate, and the presence of nonlinearity, GARCH, and level effects in its volatility. The empirical results indicate that the nonlinear asymmetric models are better than the existing models in forecasting the future level and volatility of interest rate changes. © 2006 Wiley Periodicals, Inc. Jrl Fut Mark 26:869–894, 2006     

CREATING ORDER FORECASTS: POINT‐OF‐SALE OR ORDER HISTORY?

Point‐of‐sale (POS) data, shared by retailers, is often touted as the solution to suppliers' ongoing challenge of accurate order forecasting. However, we find neither empirical evidence of increased order forecast accuracy from the literature, nor consistent use of POS data in suppliers' order forecasting processes. Using a sample containing weekly POS and order data for 10 ready‐to‐eat (RTE) cereal stock‐keeping‐units (SKU's), 7 yogurt SKU's, and 7 canned soup SKU's from 18 retailer distribution centers (DC's) throughout the U.S, our research compares historical POS and order data as order forecasting inputs and finds that POS data does not always outperform order data in terms of order forecast accuracy. While we did find that POS data is a better forecast input in a majority of the forecasts and that on average POS data produces a lower order forecast error, we find that there remain a large number of forecasts where order data is a better predictor than is POS data. Hence, we operationalize this comparison in terms of the frequency and magnitude of order forecast improvement based on POS data. We then hypothesize affecting factors and empirically test these relationships.     

我国期货市场的波动性与有效性--基于三大交易市场的实证分析

通过对中国三大期货市场的铜、黄豆和小麦三种主要期货品种收益率的分布与波动性的实证分析 ,论证了其时间序列存在ARCH效应 ;运用GARCH模型对这三种期货品种进行了拟合分析和统计检验 ,检验结果表明这三个期货品种的波动性均具有很高的持续性 ,但大连黄豆的波动持续性弱于上海铜和郑州小麦 ,其波动性受各种外部冲击的影响较大 ;通过GARCH( 1 ,1 )的市场有效性检验 ,论证了中国期货市场尚未达到弱式有效 ,市场风险较大。     

The information content of implied volatility: Evidence from Australia

This study develops an implied volatility index for the Australian stock market, termed as the AVX, and assesses its information content. The AVX is constructed using S&P/ASX 200 index options with a constant time‐to‐maturity of three months. It is observed that the AVX has a significant negative and asymmetric relationship with S&P/ASX 200 returns. When evaluating the forecasting power of the AVX for future stock market volatility, it is found that the AVX contains important information both in‐sample and out‐of‐sample. In‐sample, the AVX significantly improves the fit of a GJR‐GARCH(1, 1) model. Out‐of‐sample, the AVX significantly outperforms the RiskMetrics approach and the GJR‐GARCH(1, 1) model, with its highest forecasting power at the one‐month forecasting horizon. © 2009 Wiley Periodicals, Inc. Jrl Fut Mark 30:134–155, 2010     

Improving volatility prediction and option valuation using VIX information: A volatility spillover GARCH model

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).     

中国开放式基金收益及其波动性的周内效应研究

了解基金收益及其波动性是否存在周内效应对投资者非常重要,投资者可以利用收益及其波动性的变动信息调整投资组合,增加投资收益。运用均值方程含有虚拟变量的GARCH(1,1)模型和条件方差方程含有虚拟变量的修正的GARCH(1,1)模型,我们分别对2003年6月1日至2005年8月18日期间中国开放式基金收益的周内效应和收益波动性的周内效应进行实证研究,结果显示,在研究期间内样本基金收益及收益的波动性在周三这一天显著不同于其他交易日,即存在“周三效应”。     

中国线材期货价格指数波动率预测能力分析

利用上海期货交易所线材期货15分钟高频价格数据构造已实现波动率估计序列,并以此作为参考标准,运用6种损失函数以及Diebold-Mariano检验法检验4类不同波动率模型对线材期货价格波动的样本外预测能力,显示,中国线材期货市场,基于高频数据的GJR(1,1)模型具有最为出色的波动率预测能力,而在某些损失函数标准下,HYGARCH(1,d,1)与GARCH(1,1)模型也体现出了较好的波动率预测能力。     

The performance of event study approaches using daily commodity futures returns

Simulations are conducted to assess the inferential accuracy of statistical event study approaches using daily futures returns. Methods examined include constant mean return models and several regression models—OLS, GARCH(1,1), and a GARCH(1,1) model having an error term with a Student's t distribution. The simulations address four of the most commonly analyzed agricultural futures commodities—corn, soybeans, live cattle, and hogs. In terms of the size of the test statistics, constant mean return models with short normal periods perform poorly, leading to unacceptably high rejection rates of the null hypothesis. Test statistics from constant mean return models with longer normal periods, OLS, and GARCH specifications provide rejection rates largely consistent with those of a unit normal distribution. Test statistics from all models are powerful enough to detect abnormal performance levels below those that would trigger limit locks. At small levels of abnormal performance the GARCH(1,1) model with a t distribution was consistently the most powerful model. © 2004 Wiley Periodicals, Inc. Jrl Fut Mark 24:533–555, 2004     

Time‐varying jump risk premia in stock index futures returns

This study tests the presence of time‐varying risk premia associated with extreme news events or jumps in stock index futures return. The model allows for a dynamic jump component with autoregressive jump intensity, long‐range dependence in volatility dynamics, and a volatility in mean structure separately for the normal and extreme news events. The results show significant jump risk premia in four stock market index futures returns including the DAX, FTSE, Nikkei, and S&P500 indices. Our results are robust to various specifications of conditional variance including the plain GARCH, component GARCH, and Fractionally Integrated GARCH models. We also find the time‐varying risk premium associated with normal news events is not significant across all indices. © 2011 Wiley Periodicals, Inc. Jrl Fut Mark 32:639–659, 2012     

Realized Volatility Forecast: Structural Breaks,Long Memory,Asymmetry, and Day‐of‐the‐Week Effect

We investigate the properties of the realized volatility in Chinese stock markets by employing the high‐frequency data of Shanghai Stock Exchange Composite Index and four individual stocks from Shanghai Stock Exchange and Shenzhen Stock Exchange, and find that the volatility exhibits the properties of long‐term memory, structural breaks, asymmetry, and day‐of‐the‐week effect. In addition, the structural breaks only partially explain the long memory. To capture these properties simultaneously, we derive an adaptive asymmetry heterogeneous autoregressive model with day‐of‐the‐week effect and fractionally integrated generalized autoregressive conditional heteroskedasticity errors (HAR‐D‐FIGARCH) and use it to conduct a forecast of realized volatility. Compared with other heterogeneous autoregressive realized volatility models, the proposed model improves the in‐sample fit significantly. The proposed model is the best model for the day‐ahead realized volatility forecasts among the six models based on various loss functions by utilizing the superior predictive ability test.     

Are the East Asian markets integrated? Evidence from the ICAPM

We test a conditional international asset pricing model with both world market and domestic risk included as independent pricing factors for five East Asian markets, the US and World markets. We model second moments and risk exposures using a bi-diagonal multivariate GARCH(1,1) process. We document that this novel GARCH specification provides a significantly better fit of the return process than a standard diagonal specification. Although exposure to world market risk carries a significant premium across all markets, we find little support for the hypothesis that exposure to residual country risk is rewarded. However, residual country returns are significantly related to exchange rate changes. Hence, we find surprisingly little evidence of market segmentation in East Asia over the period 1985–1998.     

The predictability of carbon futures volatility: New evidence from the spillovers of fossil energy futures returns

In this paper, we find new evidence for the carbon futures volatility prediction by using the spillovers of fossil energy futures returns as a powerful predictor. The in-sample results show that the spillovers have a significantly positive effect on carbon futures volatility. From the out-of-sample analysis with various loss functions, we find that fossil energy return spillovers significantly outperform the benchmark and show better forecasting performance than the competing models using dimension reduction, variable selection, and combination approaches. The predictive ability of the spillovers also holds in long-term forecasting and does not derive from other carbon-related variables. It can bring substantial economic gains in the portfolio exercise within carbon futures. Finally, we provide economic explanations on the predictive ability of the fossil energy return spillover by the channels of the carbon emission uncertainty and the investor sentiment on the warming climate.     

A simplified approach to modeling the co‐movement of asset returns

The authors propose a simplified multivariate GARCH (generalized autoregressive conditional heteroscedasticity) model (the S‐GARCH model), which involves the estimation of only univariate GARCH models, both for the individual return series and for the sum and difference of each pair of series. The covariance between each pair of return series is then imputed from these variance estimates. The proposed model is considerably easier to estimate than existing multivariate GARCH models and does not suffer from the convergence problems that characterize many of these models. Moreover, the model can be easily extended to include more complex dynamics or alternative forms of the GARCH specification. The S‐GARCH model is used to estimate the minimum‐variance hedge ratio for the FTSE (Financial Times and the London Stock Exchange) 100 Index portfolio, hedged using index futures, and compared to four of the most widely used multivariate GARCH models. Using both statistical and economic evaluation criteria, it was found that the S‐GARCH model performs at least as well as the other models that were considered, and in some cases it was better. © 2007 Wiley Periodicals, Inc. Jrl Fut Mark 27:575–598, 2007     

上证50指数收益率特征及其波动性分析

研究国内非综合指数即成份指数(上证50指数)的收益率特征及其波动性,可以估计得出对指数风险收益具有较好预测作用的自回归———GARCH(1,1)-M模型,并实证分析指数收益的风险特性、稳定性、波动性等特征,这对当前探讨上证50指数相关指数衍生品推出及其投资分析均具有重要意义。