Te Tari Pāngarau me te Tatauranga
Department of Mathematics & Statistics

STAT352 Applied Time Series

Second Semester
18 points
Not available after 2018

An introduction to the practical aspects of the statistical analysis of time series and its application to the physical sciences and econometrics. Topics include seasonal decomposition, identification and estimation of ARIMA models, seasonal ARIMA models, and forecasting. Time series analysis is the statistical analysis of ordered sequences of data. The distinguishing feature is that these data are correlated. Such data arise in a bewildering range of application areas that include:

  • Climatology: Estimation long term changes of climate
  • Economics: Analysis of quarterly or monthly CPI, unemployment rates
  • Finance: Estimating volatility of stock market returns
  • Dendrochronology: Paleoclimate reconstruction based on series of tree ring widths
  • Sonar: Detection of underwater signals

as well as communications, speech compression, seismology, control theory, and many more.

Paper details

This paper examines a range of statistical techniques that can be used for the analysis of data that has been observed sequentially through time. Topics include seasonal adjustment, identification and estimation of ARMA and ARIMA models, intervention analysis and forecasting. Applications will be drawn from many disciplines ranging from econometrics to environ-mental monitoring.

Potential students

Suitable for students doing a Major or a Minor in Statistics.

It is also of interest to students in Economics or Finance.

Main topics

  • Classical decomposition of time series into a trend, seasonal and irregular component
  • Models for stationary time series; autoregression, moving average and ARMA models; identification, estimation and diagnostic testing; forecasting from ARMA models
  • ARIMA and seasonal ARIMA models for series with trend and seasonal components; forecasting


A course of regression methods such as STAT 241 Regression and Modelling 1 will be sufficient. ECON 210 or FINC 203 are also acceptable. There is no mathematics prerequisite.

Required text

Lecture notes will be available online.

Useful references

P.J. Brockwell and R.A. Davis (2016) Introduction to Time Series and Forecasting (3rd Edition) Springer. Available online in Library

C. Chatfield, (1996) Analysis of Time Series: An Introduction (5th Edition) Chapman and Hall.

P.J. Diggle (1990) Time series: A biostatistical introduction Oxford Science Publications.


Dr Ting Wang, Room 518


Monday and Wednesday, 1pm, room MA240, Science III building.

Thursday, 12noon, room MA240, Science III building. Thursday lectures will be on odd-numbered semester weeks only, starting in the first week: 1, 3, 5, 7, etc.

(Subject to change with notice in advance)


Thursdays 3pm, MA124 (level 1 computing lab, Science III), starting in week 2 of semester.

Internal Assessment

The internal assessment is made up from 5 assignments and 1 mid-semester test.

Exam format

A 3-hour exam containing three sections: Multiple choice, Concepts and theory, and Applications and real-world data.

Final mark

While we strive to keep details as accurate and up-to-date as possible, information given here should be regarded as provisional. Individual lecturers will confirm teaching and assessment methods.

Sunspot records were some of the data sets that stimulated research on statistical analysis of time series in the 20th century.

Here is the activity record from 1955 to 2000: