Forecasting financial time series volatility using Particle Swarm Optimization trained Quantile Regression Neural Network

Accurate forecasting of volatility from financial time series is paramount in financial decision making. This paper presents a novel, Particle Swarm Optimization (PSO)-trained Quantile Regression Neural Network namely PSOQRNN, to forecast volatility from financial time series. We compared the effectiveness of PSOQRNN with that of the traditional volatility forecasting models, i.e., Generalized Autoregressive Conditional Heteroskedasticity (GARCH) and three Artificial Neural Networks (ANNs) including Multi-Layer Perceptron (MLP), General Regression Neural Network (GRNN), Group Method of Data Handling (GMDH), Random Forest (RF) and two Quantile Regression (QR)-based hybrids including Quantile Regression Neural Network (QRNN) and Quantile Regression Random Forest (QRRF). The results indicate that the proposed PSOQRNN outperformed these models in terms of Mean Squared Error (MSE), on a majority of the eight financial time series including exchange rates of USD versus JPY, GBP, EUR and INR, Gold Price, Crude Oil Price, Standard and Poor 500 (S&P 500) Stock Index and NSE India Stock Index considered here. It was corroborated by the Diebold–Mariano test of statistical significance. It also performed well in terms of other important measures such as Directional Change Statistic (Dstat) and Theil's Inequality Coefficient. The superior performance of PSOQRNN can be attributed to the role played by PSO in obtaining the better solutions. Therefore, we conclude that the proposed PSOQRNN can be used as a viable alternative in forecasting volatility.     

A SAX-GA approach to evolve investment strategies on financial markets based on pattern discovery techniques

This paper presents a new computational finance approach, combining a Symbolic Aggregate approXimation (SAX) technique together with an optimization kernel based on genetic algorithms (GA). The SAX representation is used to describe the financial time series, so that, relevant patterns can be efficiently identified. The evolutionary optimization kernel is here used to identify the most relevant patterns and generate investment rules. The proposed approach was tested using real data from S&P500. The achieved results show that the proposed approach outperforms both B&H and other state-of-the-art solutions.     

An Efficient Stochastic Simulation Algorithm for Bayesian Unit Root Testing in Stochastic Volatility Models

In financial times series analysis, unit root test is one of the most important research issues. This paper is aimed to propose a new simple and efficient stochastic simulation algorithm for computing Bayes factor to detect the unit root of stochastic volatility models. The proposed algorithm is based on a classical thermodynamic integration technique named path sampling. Simulation studies show that the test procedure is efficient under moderate sample size. In the end, the performance of the proposed approach is investigated with a Monte Carlo simulation study and illustrated with a time series of S&P500 return data.     

Using a GPU-CPU architecture to speed up a GA-based real-time system for trading the stock market

The use of mechanical trading systems allows managing a huge amount of data related to the factors affecting investment performance (macroeconomic variables, company information, industrial indicators, market variables, etc.) while avoiding the psychological reactions of traders when they invest in financial markets. When trading is executed in an intra-daily frequency instead a daily frequency, mechanical trading systems needs to be supported by very powerful engines since the amount of data to deal with grow while the response time required to support trades gets shorter. Numerous studies document the use of genetic algorithms (GAs) as the engine driving mechanical trading systems. The empirical insights provided in this paper demonstrate that the combine use of GA together with a GPU-CPU architecture speeds up enormously the power and search capacity of the GA for this kind of financial applications. Moreover, the parallelization allows us to implement and test previous GA approximations. Regarding the investment results, we can report 870% of profit for the S&P 500 companies in a 10-year time period (1996–2006), when the average profit of the S&P 500 in the same period was 273%.     

A hybrid modeling approach for forecasting the volatility of S&P 500 index return

Forecasting volatility is an essential step in many financial decision makings. GARCH family of models has been extensively used in finance and economics, particularly for estimating volatility. The motivation of this study is to enhance the ability of GARCH models in forecasting the return volatility. We propose two hybrid models based on EGARCH and Artificial Neural Networks to forecast the volatility of S&P 500 index. The estimates of volatility obtained by an EGARCH model are fed forward to a Neural Network. The input to the first hybrid model is complemented by historical values of other explanatory variables. The second hybrid model takes as inputs both series of the simulated data and explanatory variables. The forecasts obtained by each of those hybrid models have been compared with those of EGARCH model in terms of closeness to the realized volatility. The computational results demonstrate that the second hybrid model provides better volatility forecasts.     

Modeling and simulation of the market fluctuations by the finite range contact systems

In this paper, we model a new random stock price model for the stock markets based on the finite range contact process, which is a model for epidemic spreading that mimics the interplay of local infections and recovery of individuals, it is a member of a class of stochastic processes known as interacting particle systems. Then, we analyze the statistical behaviors of Shanghai Stock Exchange (SSE) Composite Index, Shenzhen Stock Exchange (SZSE) Composite Index, Dow Jones Industrial Average Index (DJIA), Nasdaq Composite Index (IXIC), the standard and Poor’s 500 Index (S&P500) and the simulative data derived from the finite range contact model by comparison. And six individual Chinese stocks from large-cap, mid-cap and small-cap categories are discussed. Furthermore, we investigate the long range correlations of the returns for these indices and the corresponding simulative data by applying the detrended fluctuation analysis. At last, the positive part of the probability distributions of the logarithmic returns for the actual data and the simulative data are studied by the q-Gaussian dynamic systems. The main objective of this work is to discuss the impact on the returns with the different range financial models.     

Evolving Stacked Time Series Predictors with Multiple Window Scales and Sampling Gaps

We apply evolutionary programming to search for the optimal combination of stacked time series predictors with multiple window scales and sampling gaps. In this approach, the evolutionary process is ensured to proceed smoothly towards the optimal solution by using a control strategy based on the similarity level between the genotypes from two successive generations. Our experiments on both sunspots and S&P500 price index predictions demonstrate that this method significantly improves the prediction accuracy compared with the constrained least squared regression.     

On neuro-wavelet modeling

We survey a number of applications of the wavelet transform in time series prediction. The Haar à trous wavelet transform is proposed as a means of handling time series data when future data is unknown. Results are exemplified on financial futures and S&P500 data. Nonlinear and linear multiresolution autoregressionmodels are studied. Experimentally, we show that multiresolution approaches can outperform the traditional single resolution approach to modeling and prediction.     

Using Boosting for Financial Analysis and Performance Prediction: Application to S&;P 500 Companies,Latin American ADRs and Banks

This paper demonstrates how the boosting approach can support the financial analysis functions in two ways: (1) As a predictive tool to forecast corporate performance, and rank accounting and corporate variables according to their impact on performance, and (2) As an interpretative tool to generate alternating decision trees that capture the non-linear relationship among accounting and corporate governance variables that determine performance. We compare our results using Adaboost with logistic regression, bagging, and random forests. We conduct 10-fold cross-validation experiments on one sample each of S&P 500 companies, American Depository Receipts (ADRs) of Latin American companies and Latin American banks. Adaboost results indicate that large companies perform better than small companies, especially when these companies have a limited long-term assets to sales ratio. Performance improves for large LAADR companies when the country of residence is characterized by a weak rule of law. In the case of S&P 500 companies, performance increases when the compensation for top officers is mostly variable.     

Nonlinearity in Forecasting of High-Frequency Stock Returns

Using high-frequency S&P 500 data, we examined intraday efficiency by comparing the ability of several nonlinear models to forecast returns for horizons of 5, 10, 30 and 60?min. Taking into account fat tails and volatility dynamics, we compared the forecasting performance of simple random walk and autoregressive models with Markov switching, artificial neural network and support vector machine regression models in terms of both statistical and economic criteria. Our empirical results for out-of-sample forecasts for high and low volatility samples at different time periods provide weak evidence of intraday predictability in terms of statistical criteria, but corroborate the superiority of nonlinear model predictability using economic criteria such as trading rule profitability and value-at-risk calculations.