An Introduction to the Special Issue on Computational Techniques for Trading Systems, Time Series Forecasting, Stock Market Modeling, Financial Assets Modeling

The aim of the special Issue “Computational Techniques for Trading Systems, Time Series Forecasting, Stock Market Modeling, Financial Assets Modeling” is to present some of the latest research carried out in the field of computational finance. Financial time series modelling is regarded as one of the most challenging forecasting problems due to the fact that financial time series are inherently noisy, non-stationary and deterministically chaotic. This is strongly associated with the heterogeneous characteristics of traders (e.g. fundamentalist vs. technical analysts). Using the traditional computational techniques, it has been not possible to model the large proportion of noise and the changing distribution in financial data, respectively. Therefore, novel computational methods have emerged recently to bridge this gap. Various variants of GARCH models, neural networks, machine learning methods, fuzzy rulebased systems, agent-based models, and non-linear dynamical systems have been successfully applied in financial time series modelling. The importance of these computational methods grew as financial researchers and practitioners realized that additional variance in complex financial data can be explained. A large body of the literature has focused on evaluating the forecast accuracy of financial return and volatility, respectively. The purpose of the models is to model and predict the return and risk of financial assets in order to optimize the investment decision process through automated trading systems. In this editorial introduction, we provide a short overview of the papers contained in this special issue. Prof. Wei et al. investigate the leverage effect from the sector-specific point of view. Using the ARMA-GARCH model on CSI 300 sub-indices reflecting specific sectors, the empirical results of this study indicate that the GARCH (1,1) model is capable of explaining the fluctuations in the industry there are persistent. Additionally, TARCH (1,1) and EGARCH (1,1) models were used to examine the leverage effect of external factors and information asymmetry effect in various sectors. The results illustrate the existence of significant leverage effect between industries. Specifically, the so-called "Lee bad news" produce stronger fluctuations than the same amount of "good news". The leverage effect was considerably large especially in the case of consumer industry index. Prof. Huang et al. analyse liquidity risk premium in corporate bond spread. The panel data of corporate bonds from the Shenzhen Stock Exchange and Shanghai Stock Exchange were used to demonstrate that bond age is the most important determinant of liquidity risk premium. Other proxies of liquidity risk (squared price return, issued amount, volume and trading turnover) have also shown significant impact on liquidity risk premium. Specifically, liquidity risk premium was positively correlated with squared price return, bond age, and bond volume, respectively, but negatively correlated with issued amount. The random effect model employed for the analysis has shown promising results as it explained more than 30 % of the variance in liquidity risk premium. Prof. Wang et al. study the dynamics between stock index and stock index futures returns. Using 1-min high-frequency data, this paper investigates intraday return dynamics between CSI 300 and corresponding index futures. Such an approach is unique for emerging markets. Both spot and futures return series have shown stationarity and a cointegration relationship between spot price and futures price have been observed. Employing a wide range of models (VAR model, VECM, Granger causality test, variance decomposition, and impulse response function) the findings demonstrate that the newly established stock index futures markets in WSEAS TRANSACTIONS on BUSINESS and ECONOMICS Petr Hájek, Filippo Neri