一种改进的自适应滤波预测方法

本文对确定型常参数预测方法中的自适应滤波技术加以改进，即在最优化理论中对多变量函数极值问题的直接迭代搜索性解法中引入衰减／增益函数，在建立预测模型时把对时间序列分析定量方法与专家对相关因素的定性分析有机地结合起来，提高了原自适应滤波模型的解释能力和拟合水平，特别适合于存量经济变量和某些流量经济变量在发生比较急剧变化阶段的预测．作者应用此模型对处于成长期的我国蜂房式移动电话的市场需求进行了实证研究

AN IMPROVED SELF ADAPTIVE FILTER FORECASTING MODEL

This paper improved the self adaptive filter forecasting model which is one of deterministic constant parameter forecasting models. An attenuation/gain function is drawn into the direct iteration search method that belongs to the multi variables extreme value research based on the optimization theory, so that it combines the quantitative method with the qualitative analysis by the experts and increases the explanatory ability and simulation level of the original model. As an empirical study, the anthors applied the improved method to forecast the medium term demand for cellular phones in China.