Asymmetric time series and temporal aggregation
1999 (English)In: Review of Economics and Statistics, ISSN 0034-6535, E-ISSN 1539-9142, Vol. 81, no 2, 341-344 p.Article in journal (Refereed) Published
The detection of nonlinearities could depend on the sampling frequency. Asymmetric monthly series may become symmetric when aggregated to quarterly or annual frequencies. We test against nonlinearity using the nonlinear autoregressive asymmetric moving average (ARasMA) model, which nests the linear ARMA model as a special case. Using monthly, quarterly, and annual Swedish unemployment series, we find support for symmetry/linearity in the annual series but not in the monthly and quarterly series.
Place, publisher, year, edition, pages
1999. Vol. 81, no 2, 341-344 p.
Research subject Economics
IdentifiersURN: urn:nbn:se:uu:diva-68872DOI: 10.1162/003465399558120OAI: oai:DiVA.org:uu-68872DiVA: diva2:96783