Shanghai Stock Composite Index Research Articles

Shanghai Stock Composite Index Forecasts: Evidence from ARIMA and LSTM

Prediction of stock prices is a classic problem. People are always trying to predict stock prices as accurately as possible, and also trying to build stock price model. With the development of technology, machine learning has been applied more and more to stock prediction problems, and good results have been obtained. This paper selects the Shanghai Stock Composite Index from 1990 to 2016 and forecasts its closing price. First, the ARIMA (1,1,1) model is established for the closing price sequence after first-order difference, and then the two-layer LSTM model is constructed to visualize the prediction results of the two models and calculate rmse respectively. Comparing the prediction results of the two models, it is concluded that LSTM has better prediction effect on the data set used in this paper. This paper find that ARIMA has some cons that it is Not built for long-term forecasting and parameters are subjective. It does not predict as well as LSTM when there is a turning point.

The Asymmetry of Shanghai Composite Index Volatility—Stochastic Volatility Models Based on GHST Distribution

In this paper, we analyzed how the asymmetric stochastic volatility models with GHST distribution capture the asymmetry of stock index volatility in China. Under the setting of fat-tail distribution, we introduced the correlation parameter ρ of two error terms to refine the classification of ASV model from two aspects of Contemporaneous correlation and Subsequent correlation. So we could compare the effect of ASV model in demonstrating the asymmetry of stock index volatility under the above different settings. Using the daily returns of Shanghai stock composite index, we concluded that the ASV model with GHST distribution and Subsequent correlation between error terms can better describe the asymmetry of the stock index in China. The DIC value and Kupiec test verified the adequacy and the effectiveness of risk measurement of the above model respectively.

TIME-VARYING LONG MEMORIES OF THE CHINESE CURRENCY AND STOCK MARKETS BASED ON THE HURST EXPONENT

Based on the daily return and volatility series of the Chinese yuan (RMB)/US dollar (USD) exchange rate and the Shanghai Stock Composite Index, the time-varying long memories of the Chinese currency and stock markets are investigated by comprehensively using the rescaled range (R/S), the modified R/S, and the detrended fluctuation analysis methods. According to the results drawn: (1) the efficiency of the Chinese currency market has not improved significantly, whereas the efficiency of the Chinese stock market has improved steadily, (2) volatility series presents longer memory than return series either in the Chinese currency or stock market and (3) the time-varying Hurst exponent of the Chinese currency market is sensitive to the reform that enhances the flexibility of the RMB exchange rate. Moreover, we find that short-term bidirectional Granger causal relationship exists, but no long-run equilibrium relationship between the time-varying Hurst exponents of the Chinese currency and stock markets was found based on the Granger causality and cointegration tests, respectively.

Multifractal detrended cross-correlations between the Chinese exchange market and stock market

Based on the daily price data of the Chinese Yuan (RMB)/US dollar exchange rate and the Shanghai Stock Composite Index, we conducted an empirical analysis of the cross-correlations between the Chinese exchange market and stock market using the multifractal cross-correlation analysis method. The results demonstrate the overall significance of the cross-correlation based on the analysis of a statistic. Multifractality exists in cross-correlations, and the cross-correlated behavior of small fluctuations is more persistent than that of large fluctuations. Moreover, using the rolling windows method, we find that the cross-correlations between the Chinese exchange market and stock market vary with time and are especially sensitive to the reform of the RMB exchange rate regime. The previous reduction in the flexibility of the RMB exchange rate in July 2008 strengthened the persistence of cross-correlations and decreased the degree of multifractality, whereas the enhancement of the flexibility of the RMB exchange rate in June 2010 weakened the persistence of cross-correlations and increased the multifractality. Finally, several relevant discussions are provided to verify the robustness of our empirical analysis.

Modeling dependence based on mixture copulas and its application in risk management

This paper is concerned with the statistical modeling of the dependence structure of multivariate financial data using the copula, and the application of copula functions in VaR valuation. After the introduction of the pure copula method and the maximum and minimum mixture copula method, authors present a new algorithm based on the more generalized mixture copula functions and the dependence measure, and apply the method to the portfolio of Shanghai stock composite index and Shenzhen stock component index. Comparing with the results from various methods, one can find that the mixture copula method is better than the pure Gaussian copula method and the maximum and minimum mixture copula method on different VaR level.

Pricing bivariate option under GARCH-GH model with dynamic copula: application for Chinese market

This paper develops the method for pricing bivariate contingent claims under general autoregressive conditionally heteroskedastic (GARCH) process. In order to provide a general framework being able to accommodate skewness, leptokurtosis, fat tails as well as the time-varying volatility that are often found in financial data, generalized hyperbolic (GH) distribution is used for innovations. As the association between the underlying assets may vary over time, the dynamic copula approach is considered. Therefore, the proposed method proves to play an important role in pricing bivariate option. The approach is illustrated for Chinese market with one type of better-of-two markets claims: call option on the better performer of Shanghai Stock Composite Index and Shenzhen Stock Composite Index. Results show that the option prices obtained by the GARCH-GH model with time-varying copula differ substantially from the prices implied by the GARCH-Gaussian dynamic copula model. Moreover, the empirical work displays the advantage of the suggested method.

A new algorithm based on copulas for VaR valuation with empirical calculations

This paper concerns the application of copula functions in VaR valuation. The copula function is used to model the dependence structure of multivariate assets. After the introduction of the traditional Monte Carlo simulation method and the pure copula method we present a new algorithm based on mixture copula functions and the dependence measure, Spearman’s rho. This new method is used to simulate daily returns of two stock market indices in China, Shanghai Stock Composite Index and Shenzhen Stock Composite Index, and then empirically calculate six risk measures including VaR and conditional VaR. The results are compared with those derived from the traditional Monte Carlo method and the pure copula method. From the comparison we show that the dependence structure between asset returns plays a more important role in valuating risk measures comparing with the form of marginal distributions.

Application of support vector machines regression in prediction Shanghai stock composite index

The SVMs for regression is used to forecast Shanghai stock composite index (SSCI). Implementing structural risk minimization principle, SVMs can overcome the over-fitting problem. The regression uses ɛ-insensitive loss function. The training of SVMs leads to a quadratic programming problem and it has a global unique solution. The experiment uses BP neural networks as benchmark for comparison. The results demonstrate that the prediction figure of SSCI can help to find timing for buy or sell, the forecasting variation of SVMs is smaller than that of BP, and the direction forecasting of SVMs is more accurate than that of BP.