一种周期时间序列的预测算法

时间序列分析作为现代数据分析处理的有效方法之一,目前广泛地应用于商业、金融、证券、电信等领域,由于人们生活及消费模式的自然规律性,时间序列往往呈现出明显的周期变化趋势。论文依据移动通信网管测量数据,提出了一种针对周期时间序列的线性预测算法——自循环算法,实验表明预测精度高于传统的线性预测分析方法。

A Method for Periodical Time Series Forecast

On the grounds that the analysis of the time series has become relatively effective solutions in the data analysis and processing,and it is widely used in the field of commerce,finance,securities business and telecommunication.Because the mode of the people's life and consumption is natural and regular,the time series sometimes presents obvious periodical characteristic.On the basis of research concerning measurement data in the management of mobile communication networks,a method of linear prediction with self-circulation pattern for periodical time series is announced and it's precision is better than other traditional arithmetic.