کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
417482 | 681529 | 2013 | 11 صفحه PDF | دانلود رایگان |
The use of filters for the seasonal adjustment of data generated by the UK new car market is considered. UK new car registrations display very strong seasonality brought about by the system of identifiers in the UK registration plate, which has mutated in response to an increase in the frequency with which the identifier changes, while it also displays low frequency volatility that reflects UK macroeconomic conditions. Given the periodogram of the data, it is argued that an effective seasonal adjustment can be performed using a Butterworth lowpass filter. The results of this are compared with those based on adjustment using X-12 ARIMA and model-based methods.
► UK new car registrations display strong and changing seasonality.
► We compare adjustment using Butterworth lowpass filter, X-12 ARIMA and TRAMO-SEATS.
► These are applied to the whole sample and to each regime separately.
► All filters pick out the main trends in the sample.
► X-12 ARIMA and TRAMO-SEATS perform better when the sample was split.
Journal: Computational Statistics & Data Analysis - Volume 58, February 2013, Pages 4–14