Options
Time series modelling of the Singapore Consumer Price Index data
Author
So, Siong Teng
Supervisor
Cheang, Wai Kwong
Abstract
The aim of this Academic Exercise is to study the basic concepts, models and methods of time series analysis. In particular, identification, estimation, diagnostic checking and selection of autoregressive integrated moving average (ARIMA) models will be studied. The theory will then be put into practice by modelling the % change in Singapore CPI data (1961-2002).
It is also noted that time series observations may be affected by unusual events (or "outliersn) that can result in extraordinary patterns in the observations that, are not in accord with the rest of the observations. Outliers may have a large influence on various aspects of modelling such as model specification, parameter estimation and forecasting. As such, one aspect of this project is to construct a modified model that takes into account the effect of outliers. Forecasting will also be carried out using both models. Statistical software packages such as MINITAB and R will be used to support the analysis of the % change in CPI data.
It is also noted that time series observations may be affected by unusual events (or "outliersn) that can result in extraordinary patterns in the observations that, are not in accord with the rest of the observations. Outliers may have a large influence on various aspects of modelling such as model specification, parameter estimation and forecasting. As such, one aspect of this project is to construct a modified model that takes into account the effect of outliers. Forecasting will also be carried out using both models. Statistical software packages such as MINITAB and R will be used to support the analysis of the % change in CPI data.
Date Issued
2004
Call Number
QA280 So
Date Submitted
2004