Course Description

Time series analysis is a branch of probability statistics that is of great applicability. It is the statistic method that reveals the dynamic structure and law of system through dynamic data, and is widely applied in financial economy, signal processing, meteorology and hydrology, and so on. It is a optional major course for students majoring in applied mathematics, information and computing science. The study of this course can help students master basic theory and method of time series analysis, analyze statistic data independently, master basic modeling technique of and predict future movement of the system.

Course teaching objectives:

The student learning outcomes are what student would be able to know and to do on the completion of this course. In details are:

1. Master basic theory and method of time series analysis, be able to analyze statistical data independently, master basic modeling technique and finally, be able to predict future movement of the system.

2. Understand stability of time seires.

3. Master modeling approach of autoregression AR(p) model, moving average MA(q) model, and autoregression moving average ARMA(p,q) model.

4. Master estimation of average aotoconvariance function, predictiong through time series, and estimation of parameters in ARMA model.

Pre-course:

Mathematical Analysis, Linear Algebra, Probability and Mathematical Statistics

Textbooks and References:

Textbook:

Applied Time Series Analysis. He ShuYuan, Beijing University Press, 2009, Edition 5

References:

1. Wang ZhenLong. Applied Time Series Snalysis,China Statistics Press, 2002.

2. James D. Hamilton, Translated by Liu MingZhi. Time Series Analysis,China Social Sciences Press, 1999.

3. Peter J. Brockwell，Richard A. Davis. Translated by Tian Zheng：Theory and Methods of Time Series Analysis, Higher Education Press, 2001

4. Pan HongYu. Applied Time Series Analysis. Beijing foreign economic and trade university press, 2006

5. P. Box，M. Jenkins，C. Reinsel. Translated by Gu Lan. Applied Time Series Analysis-predict and control. China Statistics Press，1997

Course Assessment:

Form	Weight (%)
Attendance and homework	40%
Final exam	60%(term paper)

Course schedule:

Chapter	Content	Period
Chapter 1	Introduction to time series	4
Chapter 2	Autoregressive model	6
Chapter 3	Moving average model、Auto-regressive moving-average model	6
Chapter 4	Estimates of mean and the covariance function	6
Chapter 5	Using time series prediction	6
Chapter 6	The ARMA model parameters are estimated	6
Maneuver		2
Total		36

Weekly	Period	Content
Week 1	4	§Introduction to time series
Week 2	4	§Autoregressive model §The concept of partial autocorrelation function
Week 3	4	§Experiment §Moving average model
Week 4	4	§Auto-regressive moving-average model §Experiment
Week 5	4	§Estimates of mean and the covariance function §White noise test
Week 6	4	§Experiment §Using time series prediction
Week 7	4	§Forecast error、Confidence interval §Experiment
Week 8	4	§The AR(p) model parameters are estimated §The MA(p) model parameters are estimated