High-frequency Price Data Research Articles

Systemic risk monitoring model from the perspective of public information arrival

We introduce a new technique called the market correlation deviation test, which aims to detect systemic anomalies and facilitate market monitoring by using high-frequency price data from the Chinese stock market. Our method achieves a significant reduction in calculation time (to one N of the traditional method, where N represents the number of stocks) while maintaining the accuracy of calculations. This reduction is achieved by converting the covariance calculation into a difference calculation. By comparing our approach with the conventional Barndorff-Nielson-Shephard (BNS) and Corsi-Pirino-Reno (CPR) models, we demonstrate superior accuracy in identifying anomalies and efficiency in monitoring. Notably, our method can measure local price synchronization behavior, often associated with the arrival of public information, and we also observe the impact of public information on pricing. To the best of our knowledge, this study is the first to directly observe and quantify the effect of public information on pricing.

The Spillover Effects among Crude Oil Future Markets under High-Frequency Data Environment: Trivariate VAR-BEKK-GARCH Model Based on Wavelet Multiresolution Analysis

With the rapid development of financial globalization, the interconnection of international financial markets has deepened, and the relationship between domestic and international crude oil futures market price fluctuations has been increasingly strengthened. This paper utilizes 1-minute high-frequency price data of INE, WTI, and BRENT crude oil. By employing wavelet multiresolution analysis to address the ‘Calendar Effect’ present in intraday high-frequency data, and decomposing financial time series into different hierarchy of time scales. On this basis, the VAR-BEKK-GARCH model is established to study the spillover effect among the three crude oil future markets from the perspective of different trading cycles. The main conclusions of this paper are as follows: (1) The energy spectrum of wavelet multiresolution analysis shows that the price fluctuations of domestic and international crude oil futures are mainly reflected in the short-term trading cycles of 1-4 minutes; (2) The empirical results of the VAR-BEKK-GARCH model indicate that there are significant bidirectional mean and volatility spillover effects among INE, WTI, and BRENT crude oil in each trading cycle. Specifically, the mean spillover effect between INE and WTI markets shows that the short-term mean spillover effect of INE crude oil on WTI crude oil is larger, which is contrast to the medium- and long-term effect, and there is no significant pattern in the volatility spillover effect. The overall mean and volatility spillover effects of INE crude oil on BRENT crude oil are relatively larger. The overall mean and volatility spillover effects of WTI crude oil on BRENT crude oil are also relatively larger.

Multiscale Volatility Analysis for Noisy High-Frequency Prices

We present a multiscale analysis of the volatility of intraday prices from high-frequency data. Our multiscale framework includes a fractional Brownian motion and microstructure noise as the building blocks. The proposed noisy fractional Brownian motion model is shown to possess a variety of volatility behaviors suitable for intraday price processes. Algorithms for numerical estimation from time series observations are then introduced, with a new Hurst exponent estimator proposed for the noisy fractional Brownian motion model. Using real-world high-frequency price data for a collection of U.S. stocks and ETFs, we estimate the parameters in the noisy fractional Brownian motion and illustrate how the volatility varies over different timescales. The Hurst exponent and noise level also exhibit an intraday pattern whereby the the noise ratio tends to be higher near market close.

Stochastic volatility modeling of high-frequency CSI 300 index and dynamic jump prediction driven by machine learning

<abstract><p>This paper models stochastic process of price time series of $ CSI $ $ 300 $ index in Chinese financial market, analyzes volatility characteristics of intraday high-frequency price data. In the new generalized Barndorff-Nielsen and Shephard model, the lag caused by asynchrony of market information and market microstructure noises are considered, and the problem of lack of long-term dependence is solved. To speed up the valuation process, several machine learning and deep learning algorithms are used to estimate parameter and evaluate forecast results. Tracking historical jumps of different magnitudes offers promising avenues for simulating dynamic price processes and predicting future jumps. Numerical results show that the deterministic component of stochastic volatility processes would always be captured over short and longer-term windows. Research finding could be suitable for influence investors and regulators interested in predicting market dynamics based on high-frequency realized volatility.</p></abstract>

Chinese Commodity Price Evidence Study for COVID-19 Shock and Price Stickiness Model Design

Based on the high-frequency price data, this article estimates the extent of price stickiness, identifies the pricing model, and applies the micro-results to analyze the dynamic characteristics of inflation. The results show that the price moves downward steadily during the COVID-19 epidemic. Secondly, the commodity price displays low stickiness, and the pricing model shows the time-dependent pricing (TDP) model in general. Finally, the inflation inertia is negative, indicating the macro-control is effective on COVID-19 epidemic and has the feature of contradiction to the economic cycle. And inflation inertia mainly comes from food commodities, which means that the anchoring object of the policy should be food commodities during the COVID-19 pandemic.

Intraday return predictability in the cryptocurrency markets: Momentum, reversal, or both

This paper reports evidence of intraday return predictability, consisting of both intraday momentum and reversal, in the cryptocurrency market. Using high-frequency price data on Bitcoin from March 3, 2013, to May 31, 2020, it shows that the patterns of intraday return predictability change in the presence of large intraday price jumps, FOMC announcement release, liquidity levels, and the outbreak of the COVID-19. Intraday return predictability is also found in other actively traded cryptocurrencies such as Ethereum, Litecoin, and Ripple. Further analysis shows that the timing strategy based on the intraday predictors produces higher economic value than the benchmark strategy such as the always-long or the buy-and-hold. Evidence of intraday momentum can be explained in light of the theory of late-informed investors, whereas evidence of intraday reversal, which is unique to the cryptocurrency market, can be related to investors’ overreaction to non-fundamental information and overconfidence bias.

A Factor-Based Estimation of Integrated Covariance Matrix With Noisy High-Frequency Data

This article studies a high-dimensional factor model with sparse idiosyncratic covariance matrix in continuous time, using asynchronous high-frequency financial data contaminated by microstructure noise. We focus on consistent estimations of the number of common factors, the integrated covariance matrix and its inverse, based on the flat-top realized kernels introduced by Varneskov. Simulation results illustrate the satisfactory performance of our estimators in finite samples. We apply our methodology to the high-frequency price data on a large number of stocks traded in Shanghai and Shenzhen stock exchanges, and demonstrate its value for capturing time-varying covariations and portfolio allocation.

Deep Learning Forecasting in Cryptocurrency High-Frequency Trading

Like common stocks, Bitcoin price fluctuations are non-stationary and highly noisy. Due to attractiveness of Bitcoin in terms of returns and risk, Bitcoin price prediction is attracting a growing attention from both investors and researchers. Indeed, with the development of machine learning and especially deep learning, forecasting Bitcoin is receiving a particular interest. We implement and apply deep forward neural network (DFFNN) for the analysis and forecasting Bitcoin high-frequency price data. Importantly, we seek to investigate the effect of standard numerical training algorithms on the accuracy obtained by DFFNN; namely, the conjugate gradient with Powell-Beale restarts, the resilient algorithm, and Levenberg-Marquardt algorithm. The DFFNN was applied to a big dataset composed of 65,535 samples. In terms of root mean of squared errors (RMSEs), the simulation results show that the DFFNN trained with the Levenberg-Marquardt algorithm outperforms DFFNN trained with Powell-Beale restarts algorithm and DFFNN trained with resilient algorithm. In addition, the resilient algorithm is fast which suggests that it could be promising in online training and trading. The DFFNN trained with Levenberg-Marquardt algorithm is effective and easy to implement for Bitcoin high-frequency price data forecasting.

Algorithmic Pricing and Competition: Empirical Evidence from the German Retail Gasoline Market

Economic theory provides ambiguous and conflicting predictions about the association between algorithmic pricing and competition. In this paper we provide the first empirical analysis of this relationship. We study Germany’s retail gasoline market where algorithmic-pricing software became widely available by mid-2017, and for which we have access to comprehensive, highfrequency price data. Because adoption dates are unknown, we identify gas stations that adopt algorithmic-pricing software by testing for structural breaks in markers associated with algorithmic pricing. We find a large number of station-level structural breaks around the suspected time of large-scale adoption. Using this information we investigate the impact of adoption on outcomes linked to competition. Because station-level adoption is endogenous, we use brand headquarter-level adoption decisions as instruments. Our IV results show that adoption increases margins by 9%, but only in non-monopoly markets. Restricting attention to duopoly markets, we find that market-level margins do not change when only one of the two stations adopts, but increase by 28% in markets where both do. These results suggest that AI adoption has a significant effect on competition.

Bitcoin market route to maturity? Evidence from return fluctuations, temporal correlations and multiscaling effects.

Based on 1-min price changes recorded since year 2012, the fluctuation properties of the rapidly emerging Bitcoin market are assessed over chosen sub-periods, in terms of return distributions, volatility autocorrelation, Hurst exponents, and multiscaling effects. The findings are compared to the stylized facts of mature world markets. While early trading was affected by system-specific irregularities, it is found that over the months preceding April 2018 all these statistical indicators approach the features hallmarking maturity. This can be taken as an indication that the Bitcoin market, and possibly other cryptocurrencies, carry concrete potential of imminently becoming a regular market, alternative to the foreign exchange. Since high-frequency price data are available since the beginning of trading, the Bitcoin offers a unique window into the statistical characteristics of a market maturation trajectory.

Mixed interval realized variance: A robust estimator of stock price volatility

An ex post volatility estimator, called mixed interval realized variance (MIRV), is proposed. The estimator uses high-frequency price data to provide measurements robust to the idiosyncratic noise of stock markets caused by the bid-ask bounce. The theoretical properties of the new volatility estimator are illustrated and compared with those of the two canonical realized measures: realized volatility and realized range. A simulation study adds to this comparison and highlights some favorable robustness properties of the new estimator when subject to market microstructures. The main finding is that mixed interval realized variance is robust to the presence of microstructures, but inconsistent in the hypothetical ideal scenario. The empirical illustration features Australian stocks from the ASX 20 and provides evidence that for a number of stocks the mixed interval realized variance is competitive with other realized measures under predictive likelihood when it is included in a Realized GARCH model.

Volatility Forecasting: Downside Risk, Jumps and Leverage Effect

We provide empirical evidence of volatility forecasting in relation to asymmetries present in the dynamics of both return and volatility processes. Using recently-developed methodologies to detect jumps from high frequency price data, we estimate the size of positive and negative jumps and propose a methodology to estimate the size of jumps in the quadratic variation. The leverage effect is separated into continuous and discontinuous effects, and past volatility is separated into “good” and “bad”, as well as into continuous and discontinuous risks. Using a long history of the S & P500 price index, we find that the continuous leverage effect lasts about one week, while the discontinuous leverage effect disappears after one day. “Good” and “bad” continuous risks both characterize the volatility persistence, while “bad” jump risk is much more informative than “good” jump risk in forecasting future volatility. The volatility forecasting model proposed is able to capture many empirical stylized facts while still remaining parsimonious in terms of the number of parameters to be estimated.

Uma avaliação da volatilidade dos preços da soja no mercado internacional com dados de alta frequência

Neste trabalho foram avaliados os ajustes de cinco modelos para previsão da variância, utilizando-se uma série de preços de soja, uma commodity negociada na bolsa de mercadorias de Chicago (CBOT), com dados de alta frequência. Os modelos utilizados foram do tipo GARCH, FIGARCH e ARFIMA. Foi possível observar características desta série de preços de uma commodity negociada globalmente que se apresentaram inteiramente diferentes daquelas de ativos financeiros anteriormente estudados, possivelmente em virtude da característica contínua dos preços observados, induzida pela sua negociação global independente de pregões com início e fim. Foi possível concluir que a série de dados de alta frequência encerra informações adicionais às séries de dados diários, também no caso estudado de preços da soja, e que o tradicional modelo GARCH(1,1) tem um bom desempenho também no caso dos dados de alta frequência, assim como aqueles da família ARFIMA. Recomenda-se mais investigação para o caso dos modelos FIGARCH, procurando um melhor ajuste.

Large Volatility Matrix Inference via Combining Low-Frequency and High-Frequency Approaches

It is increasingly important in financial economics to estimate volatilities of asset returns. However, most of the available methods are not directly applicable when the number of assets involved is large, due to the lack of accuracy in estimating high-dimensional matrices. Therefore it is pertinent to reduce the effective size of volatility matrices in order to produce adequate estimates and forecasts. Furthermore, since high-frequency financial data for different assets are typically not recorded at the same time points, conventional dimension-reduction techniques are not directly applicable. To overcome those difficulties we explore a novel approach that combines high-frequency volatility matrix estimation together with low-frequency dynamic models. The proposed methodology consists of three steps: (i) estimate daily realized covolatility matrices directly based on high-frequency data, (ii) fit a matrix factor model to the estimated daily covolatility matrices, and (iii) fit a vector autoregressive model to the estimated volatility factors. We establish the asymptotic theory for the proposed methodology in the framework that allows sample size, number of assets, and number of days go to infinity together. Our theory shows that the relevant eigenvalues and eigenvectors can be consistently estimated. We illustrate the methodology with the high-frequency price data on several hundreds of stocks traded in Shenzhen and Shanghai Stock Exchanges over a period of 177 days in 2003. Our approach pools together the strengths of modeling and estimation at both intra-daily (high-frequency) and inter-daily (low-frequency) levels.

Moving Average-Based Estimators of Integrated Variance

We examine moving average (MA) filters for estimating the integrated variance (IV) of a financial asset price in a framework where high-frequency price data are contaminated with market microstructure noise. We show that the sum of squared MA residuals must be scaled to enable a suitable estimator of IV. The scaled estimator is shown to be consistent, first-order efficient, and asymptotically Gaussian distributed about the integrated variance under restrictive assumptions. Under more plausible assumptions, such as time-varying volatility, the MA model is misspecified. This motivates an extensive simulation study of the merits of the MA-based estimator under misspecification. Specifically, we consider nonconstant volatility combined with rounding errors and various forms of dependence between the noise and efficient returns. We benchmark the scaled MA-based estimator to subsample and realized kernel estimators and find that the MA-based estimator performs well despite the misspecification.

A Note on the Efficient Estimation of Inflation in Brazil

This paper investigates the use of trimmed-mean estimators and time-series averaging as techniques for improving the signal-to-noise ratio in high-frequency price data. We show that trimmed-mean estimators substantially increase the efficiency of the aggregate estimator compared to the more standard mean-measures. In this way, these estimators also reduce a central bank's need to time-series average the monthly inflation estimates which greatly improves the timeliness of the inflation statistic. In the case of Brazil, we find that asymmetrically trimming 24 percent from the tails of the price-change distribution reduces the RMSE of the monthly inflation statistic as a measure of the inflation trend by 23 percent, making it as accurate as the 3-month average growth rate of the mean retail price measure. We also demonstrate that a 3-month lagged moving average of the optimal (asymmetrically) trimmed mean is as efficient an estimator of the 24-month centered moving-average retail price growth trend as the mean inflation rate averaged over any horizon.

Moving Average-Based Estimators of Integrated Variance

We examine moving average (MA) filters for estimating the integrated variance of a financial asset price in a framework where high frequency price data are contaminated with marketmicrostructure noise. We show that the sum of squared MA residuals needs to be scaled for it to be a suitable estimator of integrated variance. The scaled estimator is shown to be consistent, first-order efficient, and asymptotically Gaussian distributed about the integrated variance under restrictive assumptions. Under more plausible assumptions, such as timevarying volatility, the MA model is misspecified. This motivates an extensive simulation study of the merits of the MA-based estimator under mispecification. Specifically we consider: non-constant volatility combined with rounding errors and various forms of dependence between the noise and efficient returns. We benchmark the scaled MA-based estimator to subsample and realized kernel estimators and find that the MA-based estimator performs well despite the misspecification.

A Realized Variance for the Whole Day Based on Intermittent High-Frequency Data

We consider the problem of deriving an empirical measure of daily integrated variance (IV) in the situation where high-frequency price data are unavailable for part of the day. We study three estimators in this context and characterize the assumptions that justify their use. We show that the optimal combination of the realized variance and squared overnight return can be determined, despite the latent nature of IV, and we discuss this result in relation to the problem of combining forecasts. Finally, we apply our theoretical results and construct four years of daily volatility estimates for the 30 stocks of the Dow Jones Industrial Average. Copyright 2005, Oxford University Press.

A Realized Variance for the Whole Day Based on Intermittent High-Frequency Data

We consider the problem of deriving an empirical measure of daily integrated variance (IV) in the situation where high-frequency price data are unavailable for part of the day. We study three estimators in this context and characterize the assumptions that justify their use. We show that the optimal combination of the realized variance and squared overnight return can be determined, despite the latent nature of IV, and we discuss this result in relation to the problem of combining forecasts. Finally, we apply our theoretical results and construct four years of daily volatility estimates for the 30 stocks of the Dow Jones Industrial Average.