PLS COMPUTATIONAL MODEL FOR PREDICTION IN TIME SERIES

Some data fluctuates rapidly in a short period of time. Classical and computational models are useful in predicting thishighly volatile data. In this study Partial Least Square Regression and computational neural network models are used to explorestock market tendency. Thirteen variables are considered to predict the daily closing prices of BSE sensex data. To evaluate theprediction ability of the models, standard error values are calculated. The results revealed that Nonparametric PLS regressionmodel is better in prediction.部分数据会在短时间内发生剧烈波动，经典模型与计算模型均可用于预测这类高波动数据。本研究采用偏最小二乘回归（Partial Least Square Regression）与计算神经网络模型，对股市走势进行探究。研究选取13个变量，用于预测孟买证券交易所敏感指数（BSE Sensex）的每日收盘价。为评估各模型的预测能力，本研究计算了标准误差值。结果显示，非参数偏最小二乘回归模型具备更优异的预测性能。