中文摘要：

阐述了时间序列分析的发展历程：早期自然科学主要应用单纯的描述性分析;差分、指数和滑动平均等基本概念推动了统计性时序分析的发展;以Fourier级数为基础的周期图方法是频域分析的核心内容;时域分析经过了从AR（2）模型、AR（4）模型→AR（s）模型＋MA（h）模型→AR-MA（s,h）模型→ARIMA模型→ARCH、ARCH族模型的漫长过程.

英文摘要：

The paper tells the development process of time series analysis.The early natural sciences mainly applied simple descriptive analysis;The basic notions,such as difference、index numbers and moving averages etc.pushed the development of statistics time series analysis;The periodogram method,which was based on Fourier series,was the key content in the frequency domain analysis;The time domain analysis method has experienced a long process,which began from AR（2）,AR（4） model to AR（s） model,combining with MA（h） model,formed ARMA（s,h） model,ARIMA model and at last turned up ARCH,the class of ARCH model.