Comparative study of stock trend prediction using time delay, recurrent and probabilistic neural networks

@article{citeulike:201525, abstract = {Three networks are compared for low false alarm stock trend predictions. Short-term trends, particularly attractive for neural network analysis, can be used profitably in scenarios such as option trading, but only with significant risk. Therefore, we focus on limiting false alarms, which improves the risk/reward ratio by preventing losses. To predict stock trends, we exploit time delay, recurrent, and probabilistic neural networks (TDNN, RNN, and PNN, respectively), utilizing conjugate gradient and multistream extended Kalman filter training for TDNN and RNN. We also discuss different predictability analysis techniques and perform an analysis of predictability based on a history of daily closing price. Our results indicate that all the networks are feasible, the primary preference being one of convenience}, author = {Saad, E. and Prokhorov, D. and Wunsch, D. }, citeulike-article-id = {201525}, doi = {http://dx.doi.org/10.1109/72.728395}, journal = {Neural Networks, IEEE Transactions on}, keywords = {ann, prediction, stock}, number = {6}, pages = {1456--1470}, posted-at = {2008-03-01 02:39:12}, priority = {3}, title = {Comparative study of stock trend prediction using time delay, recurrent and probabilistic neural networks}, url = {http://dx.doi.org/10.1109/72.728395}, volume = {9}, year = {1998} }

TY - JOUR ID - citeulike:201525 L3 - citeulike-article-id:201525 N2 - Three networks are compared for low false alarm stock trend predictions. Short-term trends, particularly attractive for neural network analysis, can be used profitably in scenarios such as option trading, but only with significant risk. Therefore, we focus on limiting false alarms, which improves the risk/reward ratio by preventing losses. To predict stock trends, we exploit time delay, recurrent, and probabilistic neural networks (TDNN, RNN, and PNN, respectively), utilizing conjugate gradient and multistream extended Kalman filter training for TDNN and RNN. We also discuss different predictability analysis techniques and perform an analysis of predictability based on a history of daily closing price. Our results indicate that all the networks are feasible, the primary preference being one of convenience IS - 6 JF - Neural Networks, IEEE Transactions on EP - 1470 TI - Comparative study of stock trend prediction using time delay, recurrent and probabilistic neural networks VL - 9 SP - 1456 KW - ann KW - prediction KW - stock AU - Saad, E AU - Prokhorov, D AU - Wunsch, D PY - 1998/// UR - http://dx.doi.org/10.1109/72.728395 ER -