基于偏最小二乘法的组合S型增长曲线预测模型与应用

Prediction model and application of combined S-type growth curves based on partial least squares

在许多时间序列预测中,常用S型曲线描述增长过程。由于S型曲线形式多样,结构不同,对同样观测值预测,结果略有差别。为了增加预测精度,提高预测的可靠性,考虑不同增长曲线预测结果存在多重共线性,该文利用偏最小二乘法,建立组合S型增长曲线,实例表明组合S型增长曲线预测结果精度高。

In many time series prediction, researchers commonly use S-type curves to describe the growth pro-cess. Because of the diversity of S-type curves and the differences in their structures, the results are slightly dif-ferent out of the same observation. In order to improve the prediction accuracy and reliability, the author, taking into consideration the prediction results of different growth curves with multicollinearity, established the combined S-type growth curves with partial least squares. The illustrations show that the prediction accuracy of the com-bined S-type growth curves is high.