EXPLANATORY NOTES


INTRODUCTION

1 This publication presents monthly statistics for new motor vehicle sales.

2 The statistics are based on the VFACTS series produced by the Federal Chamber of Automotive Industries (FCAI). VFACTS reports the numbers of new motor vehicle sales by dealers and direct sales by manufacturers throughout Australia. The ABS has been publishing these statistics since the January 2002 reference month. Previous to this, the ABS published New Motor Vehicle Registrations. For further information, refer to the Information Paper: Developments in New Motor Vehicle Statistics (cat. no. 9313.0).

3 From the August 2004 reference month, sports utility vehicles (SUVs) were introduced as a separate category in the original, seasonally adjusted and trend estimates. Data on SUVs are available back to January 1994 in the time series tables Sales of New Motor Vehicles, Electronic Delivery (cat. no. 9314.0.55.001). In releases of Sales of New Motor Vehicles, Australia (cat. no. 9314.0) prior to August 2004, SUVs were included in the 'Other vehicles' category.


SCOPE AND COVERAGE

4 VFACTS data are based on retail sales of new vehicles by all FCAI members. The data are provided electronically to FCAI on a daily basis. The scope of the collection:

includes:
• passenger vehicles
• four wheel drive vehicles
• vans
• trucks
• buses
• vehicles with diplomatic and consular plates
• state/territory and commonwealth owned vehicles
• vehicles belonging to the defence forces
  excludes:
• motorcycles
• plant and equipment
• unpowered vehicles

SEASONAL ADJUSTMENT

5 Seasonally adjusted and trend estimates are produced for 'Passenger vehicles', 'Sports utility vehicles', 'Other vehicles' and 'Total vehicles' for each state and territory, and are aggregated to obtain totals for Australia.

6 Seasonal adjustment is a means of removing the estimated effects of normal seasonal variation from the original time series estimates. This is done in order to reveal the underlying non-seasonal features of the series. Seasonally adjusted estimates are derived by estimating and removing systematic calendar related effects only, such as seasonal and trading day influences, from the original estimates. Seasonal adjustment does not aim to remove the irregular or non-seasonal influences from the original series, such as the impact of the introduction of new models or industrial disputes.

7 As a general rule, 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.

8 As happens with all ABS seasonally adjusted series, the seasonal adjustment factors are reviewed and revised on a regular basis, taking into account the latest information in relation to features in the time series. For further information about seasonal adjustment, please refer to An Introductory Course in Time Series Analysis - Electronic Delivery (cat. no. 1346.0.55.001). Further advice on revisions will be made in future releases of this publication.

9 Since the November 2003 reference month, the method of calculating seasonally adjusted estimates changed from Forward Factor to Concurrent. The Forward Factor method used annual seasonal reanalysis of the original time series estimates to derive seasonal adjustment factors for the forthcoming year. Under this method, the projected seasonal factors, or forward factors, were not updated until the next annual seasonal reanalysis. The Concurrent method re-estimates seasonal factors as each new data point becomes available. The Concurrent method eliminates the need for projected seasonal factors.

10 More recently, the ABS has implemented improved methods of producing seasonally adjusted estimates, focused on the application of Autoregressive Integrated Moving Average (ARIMA) modelling techniques. ARIMA modelling is a technique that can be used to extend original estimates beyond the end of a time series. The use of ARIMA modelling generally results in a reduction in revisions to the seasonally adjusted estimates when subsequent data becomes available. The New Motor Vehicles collection uses ARIMA modelling where appropriate for individual time series. Factors such as length and volatility of a time series affect whether ARIMA modelling is appropriate. ARIMA models are assessed as part of the annual seasonal reanalysis and following a reanalysis of data up to October 2007, the vast majority of New Motor Vehicle Sales series use an ARIMA model. For more information on the details of ARIMA modelling see the feature article 'Use of ARIMA modelling to reduce revisions' in the October 2004 issue of Australian Economic Indicators (cat. no. 1350.0).


TREND ESTIMATES

11 In highly volatile time series, fluctuations that are neither systematic nor predictable, can make it difficult to interpret the underlying movement of the series even after adjustment for seasonal variation. The smoothing of seasonally adjusted estimates to produce trend estimates reduces the impact of the volatile component of the seasonally adjusted series. Trend estimates are derived by applying a 13-term Henderson-weighted moving average to the respective seasonally adjusted estimates. These trend estimates are used to analyse the underlying behaviour of the series over time.

12 This smoothing technique enables trend estimates to be produced for the latest month. However, as data for subsequent months become available, the trend estimates for the most recent months are revised. Generally, subsequent revisions become smaller and after three months, usually have a negligible impact on the series. Changes in the original data and revisions to seasonal factors may also lead to revisions to the trend. As a result of the introduction of The New Tax System, a break in the monthly trend series has been inserted between June 2000 and July 2000. Care should therefore be taken in comparing the series over time. For further information, refer to Information Paper: A Guide to Interpreting Time Series - Monitoring Trends, 2003 (cat. no. 1349.0).


FURTHER INFORMATION

13 For a more detailed breakdown 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).