Short-term Stock Price Forecasting Model Based On Grey Relational Analysis And NMLPNN Optimized By GSA

Stock market prediction is regarded as a challenging task in financial time-series forecasting and the analysis and modeling of finance time series is an important task for guiding investors’ decisions and trades. Time series forecasting has been widely used to estimate future prices of stocks. Nonetheless, the prediction of prices by means of a time series is not trivial and it requires a thorough analysis of indexes, variables and other data. In addition, in a dynamic environment such as the stock market, the non-linearity of the time series is a pronounced characteristic, and this immediately affects the efficacy of stock price forecasts. In this paper, the novel integrated approaches, combining grey relational analysis(GRA), gravitational search algorithm(GSA) and the nonlinear multi-layer perceptron neural network(NMLPNN), were proposed for the short-term prices of stocks forecasting. At first, we use grey relational analysis to determine factors which have most influence on stock prices. Then, we use GSA to optimize the weights and biases of the NMLPNN. Finally, the NMLPNN, which is optimized by the GSA, is utilized to deal with the irregularity and volatility of the prices of stocks. These integrated methods are applied to forecast daily stock price data from GEI(growth enterprise index) and SPD Bank(Shanghai Pudong Development Bank) shares. By comparison of the obtained experimental results, the proposed grey relational analysis combination of GSA-NMLPNN integrated method indicates the superiority and promising performance and has a good robustness.