|Page tools: Print Page Print All|
5 Seasonal adjustment is a means of removing the estimated effects of normal seasonal variation from the original time series so that the effects of other influences can be more clearly recognised. It does not remove the effect of irregular or non-seasonal influences (e.g. the introduction of new models, industrial disputes) from the series. Irregular influences that are highly volatile can make it difficult to interpret the movement of the series even after adjustment for seasonal variation.
6 Extreme care should be exercised in using the seasonally adjusted series for new motor vehicle sales in Tasmania, the Northern Territory and the Australian Capital Territory. The small numbers and volatile nature of these data makes reliable estimation of the seasonal pattern very difficult.
7 Seasonally adjusted series are calculated for the 'Passenger vehicles', 'Other vehicles' and 'Total vehicles' series for each state and territory and are aggregated to obtain the total for Australia.
8 From the November 2003 reference month, the way in which seasonally adjusted and trend estimates are calculated was changed from the Forward Factor to the Concurrent method. The Forward Factor method relied on an annual seasonal reanalysis of the original time series estimates to derive seasonal factors that were to be applied in the forthcoming twelve months. Under this method, the projected seasonal factors, or forward factors, were not updated until the next annual seasonal reanalysis. The Concurrent method uses the most up to date original time series estimates available at each reference period to rederive seasonal factor estimates. The Concurrent method eliminates the need to use projected seasonal factors.
9 The smoothing of seasonally adjusted series to create trend estimates reduces the impact of the irregular component of the seasonally adjusted series. The trend estimates are derived by applying a 13-term Henderson-weighted moving average to the respective seasonally adjusted series. These trend series are used to analyse the underlying behaviour of the series over time.
10 While this smoothing technique enables trend estimates to be produced for the latest month, it does result in revisions to the trend estimates for the most recent months as data for subsequent months become available. Generally, subsequent revisions become smaller and after three months, usually have a negligible impact on the series. Changes in the original data and re-estimation of seasonal factors may also lead to revisions to the trend. For further information, refer to Information Paper: A Guide to Interpreting Time Series - Monitoring Trends, 2003, (cat. no. 1349.0).
11 For a more detailed break-down of the original monthly figures presented here, inquiries should be made to the Manager, VFACTS, Federal Chamber of Automotive Industries on (03) 9829 1234. Annual data on total vehicle registrations are published in Motor Vehicle Census, Australia (cat. no. 9309.0).
These documents will be presented in a new window.