رسائل دكتوراة
English
2015
Evolving Stock Market Prediction Models Using Soft Computing Techniques
Alaa F. Sheta, Sara Elsir Mohamed Ahmad Haimoura
كلية علوم الحاسوب وتقانة المعلومات-جامعة السودان للعلوم والتكنولوجيا · السودان
Stock market prediction is one of the hottest field of research lately due to its business applicationsowing to high stakes and the kinds of attractive benefits that it has to offer. The stockmarket is a dynamic, non-linear, complex, and chaotic in nature, forecasting stock marketprice is an important financial problem that is receiving increasing attention. During the lastfew years, a number of many models were presented to develop a relationship between theattributes which affect the stock index and its values. This thesis proposes a soft computingtechnique enhanced decision in financial management. The decision allows investors tomaximize their expected return while practicing the prediction against financial risks. Theimportance of the research stems from the fact that it can be used to reduce the risk associatedwith uncertainty price movements in the stock market. The literature review shows that thereare a large number of studies trying to forecast movements in the stock market, but thereis a lack of literature trying to improve stock market risk management strategies with softcomputing techniques. This thesis addresses this gap by applying the existing body of literaturein stock index forecasting with soft computing techniques to the domain of forecastingindex movements. In particular, it analyses whether there is an influence features of stocksused to predict movements of the stock index can improve forecasting the stock index movementoff an investor faces use S&P 500 dataset and data related to it to create new datasethas impact in the price; The S&P 500, or the Standard & Poor’s 500, is an American stockmarket index. The S&P 500 presented its first stock index in the year 1923. S&P 500 indexfound to have 27 influence features which affect the index values. A new market forecastingmodel based on soft computing and especially genetic programming is developed to enhancethe investor decision. The model compare with traditional model Auto regression model andivArtificial Neural Network model. The system analysis stock market and futures data andmakes a prediction about expected stock market conditions one day and next week. Selectednew Features dataset by used GA to decrease the complexity. We were expecting that notall these features are significant in computing the index. Thus, we split our work to twophases. The first phase is to develop models based on these 27 features using Multiple LinearRegression (MLR) and ANN. Not only that, but we also explored a promising technique,multigene symbolic regression Genetic Programming to provide a mathematical nonlinearrelationship between these attributes. GP found to be a powerful algorithm for providingmathematical modeling. In the second phase of this thesis, we adopted GAs as a mechanismto select the best features that contribute to the modeling process. The set of best featuresare once again used to build S & P 500 prediction models. Although the developed modelsin the second phase are with slightly less performance than in the first phase but the modelsare much simplex in complexity. This suggests that the stock market can be forecast usingsoft computing technique. Overall, this thesis concludes that the proposed model achieves asignificant improvement in the prediction of stock market index.
