Finite sample accuracy and choice of sampling frequency in integrated volatility estimation

We consider the properties of three estimation methods for integrated volatility, i.e. realized volatility, Fourier, and wavelet estimation, when a typical sample of high-frequency data is observed. We employ several different generating mechanisms for the instantaneous volatility process, e.g. Ornstein–Uhlenbeck, long memory, and jump processes. The possibility of market microstructure contamination is also entertained using models with bid-ask bounce and price discreteness, in which case alternative estimators with theoretical justification under market microstructure noise are also examined. The estimation methods are compared in a simulation study which reveals a general robustness towards persistence or jumps in the latent stochastic volatility process. However, bid-ask bounce effects render realized volatility and especially the wavelet estimator less useful in practice, whereas the Fourier method remains useful and is superior to the other two estimators in that case. More strikingly, even compared to bias correction methods for microstructure noise, the Fourier method is superior with respect to RMSE while having only slightly higher bias. A brief empirical illustration with high-frequency GE data is also included.     

Two-step estimation of the volatility functions in diffusion models with empirical applications

In this article, we develop a two-step estimation procedure for the volatility function in diffusion models. We firstly estimate the volatility series at sampling time points based on high-frequency data. Then, the volatility function estimator can be obtained by using the kernel smoothing method. The resulting estimators are presented based on high-frequency data, and are shown to be consistent and asymptotically normal. We also consider boundary issues and then propose two methods to handle them. The asymptotic normality of two boundary-corrected estimators is established under some suitable conditions. The proposed estimators are illustrated by Monte Carlo simulations and real data.     

Measuring relative volatility in high-frequency data under the directional change approach

We introduce a new approach in measuring relative volatility between two markets based on the directional change (DC) method. DC is a data-driven approach for sampling financial market data such that the data are recorded when the price changes have reached a significant amplitude rather than recording data under a predetermined timescale. Under the DC framework, we propose a new concept of DC micro-market relative volatility to evaluate relative volatility between two markets. Unlike the time-series method, micro-market relative volatility redefines the timescale based on the frequency of the observed DC data between the two markets. We show that it is useful for measuring the relative volatility in micro-market activities (high-frequency data).     

Leverage and Volatility Feedback Effects in High-Frequency Data

We examine the relationship between volatility and past andfuture returns using high-frequency aggregate equity index data.Consistent with a prolonged "leverage" effect, we find the correlationsbetween absolute high-frequency returns and current and pasthigh-frequency returns to be significantly negative for severaldays, whereas the reverse cross-correlations are generally negligible.We also find that high-frequency data may be used in more accuratelyassessing volatility asymmetries over longer daily return horizons.Furthermore, our analysis of several popular continuous-timestochastic volatility models clearly points to the importanceof allowing for multiple latent volatility factors for satisfactorilydescribing the observed volatility asymmetries.     

Forecasting financial market volatility: Sample frequency vis-à-vis forecast horizon

This paper explores the return volatility predictability inherent in high-frequency speculative returns. Our analysis focuses on a refinement of the more traditional volatility measures, the integrated volatility, which links the notion of volatility more directly to the return variance over the relevant horizon. In our empirical analysis of the foreign exchange market the integrated volatility is conveniently approximated by a cumulative sum of the squared intraday returns. Forecast horizons ranging from short intraday to 1-month intervals are investigated. We document that standard volatility models generally provide good forecasts of this economically relevant volatility measure. Moreover, the use of high-frequency returns significantly improves the longer run interdaily volatility forecasts, both in theory and practice. The results are thus directly relevant for general research methodology as well as industry applications.     

Pricing under rough volatility

From an analysis of the time series of realized variance using recent high-frequency data, Gatheral et al. [Volatility is rough, 2014] previously showed that the logarithm of realized variance behaves essentially as a fractional Brownian motion with Hurst exponent H of order 0.1, at any reasonable timescale. The resulting Rough Fractional Stochastic Volatility (RFSV) model is remarkably consistent with financial time series data. We now show how the RFSV model can be used to price claims on both the underlying and integrated variance. We analyse in detail a simple case of this model, the rBergomi model. In particular, we find that the rBergomi model fits the SPX volatility markedly better than conventional Markovian stochastic volatility models, and with fewer parameters. Finally, we show that actual SPX variance swap curves seem to be consistent with model forecasts, with particular dramatic examples from the weekend of the collapse of Lehman Brothers and the Flash Crash.     

Stock volatility predictability in bull and bear markets

The recent literature on stock return predictability suggests that it varies substantially across economic states, being strongest during bad economic times. In line with this evidence, we document that stock volatility predictability is also state dependent. In particular, in this paper, we use a large data set of high-frequency data on individual stocks and a few popular time-series volatility models to comprehensively examine how volatility forecastability varies across bull and bear states of the stock market. We find that the volatility forecast horizon is substantially longer when the market is in a bear state than when it is in a bull state. In addition, over all but the shortest horizons, the volatility forecast accuracy is higher when the market is in a bear state. This difference increases as the forecast horizon lengthens. Our study concludes that stock volatility predictability is strongest during bad economic times, proxied by bear market states.     

Does high-frequency trading actually improve market liquidity? A comparative study for selected models and measures

The increasing volume of messages sent to the exchange by algorithmic traders stimulates a fierce debate among academics and practitioners on the impacts of high-frequency trading (HFT) on capital markets. By comparing a variety of regression models that associate various measures of market liquidity with measures of high-frequency activity on the same dataset, we find that for some models the increase in high-frequency activity improves market liquidity, but for others, we get the opposite effect. We indicate that this ambiguity does not depend only on the stock market or the data period, but also on the used HFT measure: the increase of high-frequency orders leads to lower market liquidity whereas the increase in high-frequency trades improves liquidity. We hypothesize that the observed decrease in market liquidity associated with an increasing level of high-frequency orders is caused by a rise in quote volatility.     

Order flow and volatility: An empirical investigation

We study the relationship between order flow and volatility. To this end we develop a comprehensive framework that simultaneously controls for the effects of macro announcements and order flow on prices and the effect of macro announcements on volatility. Using high-frequency 30-year U.S. Treasury bond futures data, we find a statistically and economically significant relationship between the absolute value of order flow and volatility. Moreover, this relationship is robust, inter alia, to a number of factors including the introduction of liquidity effects, use of data measured over a different frequency, and market conditions.     

Risk appetite and option prices: Evidence from the Chinese SSE50 options market

This paper studies the relationship between risk appetite and the shape of the implied volatility function for SSE 50ETF options, and examines whether the effects of risk appetite differ on that of call and put options. We propose a new measure of risk appetite using high-frequency data in the stock market. Empirical results show that risk appetite has a significant impact on the level and slope of the implied volatility function, with significant differences between call and put options. In addition, we also find that risk appetite has an obvious asymmetric impact on option prices under the leverage effect.     

Aggregate volatility risk: International evidence

Using a procedure analogous to that of Ang et al. (2006), this paper documents that aggregate volatility risk does not appear to be priced in European equity markets. Specifically, based on the 2002–2016 period (for which European stock return data is available), the price of aggregate volatility risk is not statistically different from zero. Analysis based on GARCH-class and high-frequency intraday data models support these results. Consequently, contrary to what has been reported in some studies that examine U.S. data, whether aggregate volatility risk is priced in equity markets is an open question.     

Nonparametric realized volatility estimation in the international equity markets

Using high-frequency intraday data, we construct, test and model seven new realized volatility estimators for six international equity indices. We detect jumps in these estimators, construct the jump components of volatility and perform various tests on their properties. Then we use the class of heterogeneous autoregressive (HAR) models for assessing the relevant effects of jumps on volatility. Our results expand and complement the previous literature on the nonparametric realized volatility estimation in terms of volatility jumps being examined and modeled for the international equity market, using such a variety of new realized volatility estimators. The selection of realized volatility estimator greatly affects jump detection, magnitude and modeling. The properties each volatility estimator tries to incorporate affect the detection, magnitude and properties of jumps. These volatility-estimation and jump properties are also evident in jump modeling based on statistical and economic terms.     

Modeling the leverage effect with copulas and realized volatility

In this paper, we propose the use of static and dynamic copulas to study the leverage effect in the S&P 500 index. Copula models can conveniently separate the leverage effect from the marginal distributions of the return and its volatility. Daily volatility is proxied by a measure of realized volatility, which is constructed from high-frequency data. We uncover a significant leverage effect in the S&P 500 index, and this leverage effect is found to be changing over time in a highly persistent manner. Moreover the dynamic copula models are shown to outperform the static counterparts.     

Level changes in volatility models

Level shifts confound the estimation of persistence. This paper shows analytically, in simulations, and using high-frequency stock price data that models for financial volatility that feature a separate source of randomness in the volatility equation are less susceptible to this effect. Such models include recently proposed time series models for realized volatility, as opposed to GARCH models for daily observations, which are highly sensitive to unknown shifts, as has been shown before.     

Stochastic volatility: A tale of co-jumps,non-normality,GMM and high frequency data

In this article we introduce a linear–quadratic volatility model with co-jumps and show how to calibrate this model to a rich dataset. We apply GMM and more specifically match the moments of realized power and multi-power variations, which are obtained from high-frequency stock market data. Our model incorporates two salient features: the setting of simultaneous jumps in both return process and volatility process and the superposition structure of a continuous linear–quadratic volatility process and a Lévy-driven Ornstein–Uhlenbeck process. We compare the quality of fit for several models, and show that our model outperforms the conventional jump diffusion or Bates model. Besides that, we find evidence that the jump sizes are not normally distributed and that our model performs best when the distribution of jump-sizes is only specified through certain (co-) moment conditions. Monte Carlo experiments are employed to confirm this.     

Asymmetric volatility connectedness on the forex market

We show how bad and good volatility propagate through the forex market, i.e., we provide evidence for asymmetric volatility connectedness on the forex market. Using high-frequency, intra-day data of the most actively traded currencies over 2007–2015 we document the dominating asymmetries in spillovers that are due to bad, rather than good, volatility. We also show that negative spillovers are chiefly tied to the dragging sovereign debt crisis in Europe while positive spillovers are correlated with the subprime crisis, different monetary policies among key world central banks, and developments on commodities markets. It seems that a combination of monetary and real-economy events is behind the positive asymmetries in volatility spillovers, while fiscal factors are linked with negative spillovers.     

Dynamic hedging with futures: a copula-based GARCH model with high-frequency data

Modeling the joint distribution of spot and futures returns is crucial for establishing optimal hedging strategies. This paper proposes a new class of dynamic copula-GARCH models that exploits information from high-frequency data for hedge ratio estimation. The copula theory facilitates constructing a flexible distribution; the inclusion of realized volatility measures constructed from high-frequency data enables copula forecasts to swiftly adapt to changing markets. By using data concerning equity index returns, the estimation results show that the inclusion of realized measures of volatility and correlation greatly enhances the explanatory power in the modeling. Moreover, the out-of-sample forecasting results show that the hedged portfolios constructed from the proposed model are superior to those constructed from the prevailing models in reducing the (estimated) conditional hedged portfolio variance. Finally, the economic gains from exploiting high-frequency data for estimating the hedge ratios are examined. It is found that hedgers obtain additional benefits by including high-frequency data in their hedging decisions; more risk-averse hedgers generate greater benefits.     

Analysis of the intraday effects of economic releases on the currency market

Using four years of second-by-second executed trade data, we study the intraday effects of a representative group of scheduled economic releases on three exchange rates: EUR/USD, JPY/USD, and GBP/USD. Using wavelets to analyze volatility behavior, we empirically show that intraday volatility clusters increase as we approach the time of the releases, and decay exponentially after the releases. Moreover, we compare our results with the results of a poll that we conducted of economists and traders. Finally, we propose a wavelet volatility estimator which is not only more efficient than a range estimator that is commonly used in empirical studies, but also captures the market dynamics as accurately as a range estimator. Our approach has practical value in high-frequency algorithmic trading, as well as electronic market making.     

Implied volatility index for the Norwegian equity market

We introduce and evaluate the NOVIX - an implied volatility index for the Norwegian equity index OBX. NOVIX is created according to the VIX methodology. We compare the NOVIX to the German VDAX-NEW and the U.S. VIX and find that NOVIX has similar properties as these two indices. We also evaluate the VIX, VDAX-NEW and NOVIX in terms of volatility forecasting. As a benchmark model we use a precise HAR model of Corsi (2009) based on high-frequency data. All three implied volatility indices significantly improve daily, weekly and monthly forecasts of volatility of their underlying equity indices. This improvement is largest for the VIX, followed by VDAX-NEW and NOVIX.     

Combining long memory and level shifts in modelling and forecasting the volatility of asset returns

We propose a parametric state space model of asset return volatility with an accompanying estimation and forecasting framework that allows for ARFIMA dynamics, random level shifts and measurement errors. The Kalman filter is used to construct the state-augmented likelihood function and subsequently to generate forecasts, which are mean and path-corrected. We apply our model to eight daily volatility series constructed from both high-frequency and daily returns. Full sample parameter estimates reveal that random level shifts are present in all series. Genuine long memory is present in most high-frequency measures of volatility, whereas there is little remaining dynamics in the volatility measures constructed using daily returns. From extensive forecast evaluations, we find that our ARFIMA model with random level shifts consistently belongs to the 10% Model Confidence Set across a variety of forecast horizons, asset classes and volatility measures. The gains in forecast accuracy can be very pronounced, especially at longer horizons.