620-374 Sampling and Forecasting

620-202 or [01]620-204

This subject covers a range of important and generally applicable statistical methods.

Students should develop the ability to employ these methods to implement a range of practically useful statistical analyses. The following three topics will be covered:

• sample surveys: simple random sampling; stratified sampling - optimal allocation, post-stratification; cluster sampling; ratio estimation;

• time series and forecasting: patterns in time series; simple methods for exploratory data analysis; smoothing techniques; decomposition, trends and seasonal variation; simple forecasting methods; models for time series: stationarity, autocorrelation, ARMA processes; estimation and model fitting; and

• re-sampling methods: jack-knife and the bootstrap; and use of the bootstrap for exploring the sampling distribution of an estimator.

The subject develops the students' generic skills, including thinking critically and organising knowledge; analysing data, interpreting results and presenting conclusions in a clear and comprehensible manner; using computers for data analysis and presentation; and solving problems.

Up to 50 pages of written assignments during semester (20%) and a 3-hour end-of-semester written examination (80%).