Analysis of High Frequency Data in Finance:A Survey

This study examines the use of high frequency data in finance,including volatility estimation and jump tests.High frequency data allows the construction of model-free volatility measures for asset returns.Realized variance is a consistent estimator of quadratic variation under mild regularity conditions.Other variation concepts,such as power variation and bipower variation,are useful and important for analyzing high frequency data when jumps are present.High frequency data can also be used to test jumps in asset prices.We discuss three jump tests:bipower variation test,power variation test,and variance swap test in this study.The presence of market microstructure noise complicates the analysis of high frequency data.The survey introduces several robust methods of volatility estimation and jump tests in the presence of market microstructure noise.Finally,some applications of jump tests in asset pricing are discussed in this article.

A Data Transmission Method Using a Cyclic Redundancy Check and Inaudible Frequencies

This paper describes a data transmission method using a cyclic redundancy check and inaudible frequencies.The proposed method uses inaudible high frequencies from 18 k Hz to 22 k Hz generated via the inner speaker of smart devices.Using the proposed method,the performance is evaluated by conducting data transmission tests between a smart book and smart phone.The test results confirm that the proposed method can send 32 bits of data in an average of 235 ms,the transmission success rate reaches 99.47%,and the error detection rate of the cyclic redundancy check is0.53%.

Volatility-volume relationship of Chinese copper and aluminum futures market

Following Bessembinder and Seguins,trading volume is separated into expected and unexpected components.Meanwhile,realized volatility is divided into continuous and discontinuous jump components.We make the empirical research to investigate the relationship between trading volume components and various realized volatility using1min high frequency data of Shanghai copper and aluminum futures.Moreover,the asymmetry of volatility-volume relationship is investigated.The results show that there is strong positive correlation between volatility and trading volume when realized volatility and its continuous component are considered.The relationship between trading volume and discontinuous jump component is ambiguous.The expected and unexpected trading volumes have positive influence on volatility.Furthermore,the unexpected trading volume,which is caused by arrival of new information,has a larger influence on price volatility.The findings also show that an asymmetric volatility-volume relationship indeed exists,which can be interpreted by the fact that trading volume has more explanatory power in positive realized semi-variance than negative realized semi-variance.The influence of positive trading volume shock on volatility is larger than that of negative trading volume shock,which reflects strong arbitrage in Chinese copper and aluminum futures markets.

Testing long memory based on a discretely observed process

In this paper we consider the problem of testing long memory for a continuous time process based on high frequency data. We provide two test statistics to distinguish between a semimartingale and a fractional integral process with jumps, where the integral is driven by a fractional Brownian motion with long memory. The small-sample performances of the statistics are evidenced by means of simulation studies. The real data analysis shows that the fractional integral process with jumps can capture the long memory of some financial data.

Hybrid ARIMA/RBF Framework for Prediction BUX Index

In this paper, we construct and implement a new architecture and learning method of customized hybrid RBF neural network for high frequency time series data forecasting. The hybridization is carried out using two running approaches. In the first one, the ARCH (Autoregressive Conditionally Heteroscedastic)-GARCH (Generalized ARCH) methodology is applied. The second modeling approach is based on RBF (Radial Basic Function) neural network using Gaussian activation function with cloud concept. The use of both methods is useful, because there is no knowledge about the relationship between the inputs into the system and its output. Both approaches are merged into one framework to predict the final forecast values. The question arises whether non-linear methods like neural networks can help modeling any non-linearities being inherent within the estimated statistical model. We also test the customized version of the RBF combined with the machine learning method based on SVM learning system. The proposed novel approach is applied to high frequency data of the BUX stock index time series. Our results show that the proposed approach achieves better forecast accuracy on the validation dataset than most available techniques.

Empirical Evidence of the Leverage Effect in a Stochastic Volatility Model： A Realized Volatility Approach

Increasing attention has been focused on the analysis of the realized volatil- ity, which can be treated as a proxy for the true volatility. In this paper, we study the potential use of the realized volatility as a proxy in a stochastic volatility model estimation. We estimate the leveraged stochastic volatility model using the realized volatility computed from five popular methods across six sampling-frequency transaction data （from 1-min to 60- min） based on the trust region method. Availability of the realized volatility allows us to estimate the model parameters via the MLE and thus avoids computational challenge in the high dimensional integration. Six stock indices are considered in the empirical investigation. We discover some consistent findings and interesting patterns from the empirical results. In general, the significant leverage effect is consistently detected at each sampling frequency and the volatility persistence becomes weaker at the lower sampling frequency.