论文部分内容阅读
本文提出了基于自回归(AR)模型对时间序列统一建模的方法,同流行的基于自回归滑动平均(ARMA)模型对时间序列建模的方法相比,显著地减少了计算量,并且提出了 AR 模型自动辨识机,它可自动地决定时间序列所服从的AR 模型的阶和参数,因而有明显的实用价值,以太阳黑子数目的时间序列建模为例,说明了所提出的辨识的有效性。
In this paper, we present a new method to model time series based on autoregressive (AR) model. Compared with the popular autoregressive moving average (ARMA) model for time series modeling, the method reduces the computational complexity significantly and proposes The AR model automatic identification machine, which can automatically determine the order and parameter of the AR model obeyed by the time series, has obvious practical value. Taking the time series modeling of the number of sunspots as an example, it shows that the proposed identification Effectiveness.