Abstract: The paper compares the search for structural breaks in Sweden's GDP conducted with X-11 seasonally adjusted data, with seasonally unadjusted data, and with temporally aggregated data. A structural break (in 1980) is found only in the X-11 adjusted data, it is plausible to conclude that this break is due to data distortions (particularly, distortions caused by the application of the filter). However, this interpretation is only plausible a posteriori: had the seasonally unadjusted data not been available, the break found in the adjusted series could be just as well interpreted as a break in the economy and not as a break in the data. The study suggests that seasonally adjusted data should not be used when the unadjusted version is also available.

Keywords: X-11 filter; seasonal adjustment; seasonal unit roots

JEL-codes: C12; C32

18 pages, March 30, 1998

Full text files

hastef0230.fig.pdf.zip  Figures

hastef0230.fig.pdf  Figures

hastef0230.fig.ps.zip PostScript file Figures

hastef0230.fig.ps PostScript file Figures

hastef0230.pdf.zip  Full text

hastef0230.pdf  Full text

hastef0230.ps.zip PostScript file Full text

hastef0230.ps PostScript file Full text

Download statistics

• Stockholm School of Economics
• Home page for this series

Questions (including download problems) about the papers in this series should be directed to Helena Lundin ()
Report other problems with accessing this service to Sune Karlsson ().

This page generated on 2024-02-11 18:19:14.