A time series is a sequence of observations that are arranged according to the time of their outcome. The reasons of doing time series analysis are diverse, depending on the background of applications. Statisticians usually view a time series as a realization from a stochastic process. A fundamental task is to unveil the probability law that governs the observed time series. For instance, we wish to gain a better understanding of the data generating mechanism, the prediction of future values. Many recent applications of time series receive much of attention in financial areas. However, time series analysis has also exhibited its importance across many scientific areas for a long history and the desire of such analysis is still going on. In this course, our flow is to spend around two thirds of time in the treatment from a traditional look of time series then to some recent models such as GARCH, long memory models and nonparametric methods. The course level is set for master students.

A time series is a sequence of observations that are arranged according to the time of their outcome. Many recent applications of time series receive much of attention in financial areas. However, time series analysis has also exhibited its importance across many scientific areas for a long history and the desire of such analysis is still going on. This course is to provide introductory lessons in time series for the third and fourth year of undergraduates, who have major interests in statistics, and have finished some elementary courses, such as statistics, calculus, linear algebra and regression. The courses will be in three parts. The first part will be on the focus of regression, including a quick overall review of basics and an extension of regression modeling particularly required for time series data. The second part is aimed at the ARIMA models and seasonal models, which are traditional treatments established by Box and Jenkins. The third part is to introduce some new advances in time series or more real applications of time series models.

Bowerman, O'Connell, and Koehler, 2005, Forecasting, Time Series, and Regression, 4th Ed., Thomson Brooks/Cole