Premium
Another look at Swedish business cycles, 1861–1988
Author(s) -
Skalin Joakim,
Teräsvirta Timo
Publication year - 1999
Publication title -
journal of applied econometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.878
H-Index - 99
eISSN - 1099-1255
pISSN - 0883-7252
DOI - 10.1002/(sici)1099-1255(199907/08)14:4<359::aid-jae517>3.0.co;2-1
Subject(s) - business cycle , causality (physics) , pairwise comparison , econometrics , star (game theory) , economics , granger causality , constant (computer programming) , series (stratigraphy) , outcome (game theory) , mathematics , mathematical economics , keynesian economics , computer science , statistics , physics , mathematical analysis , paleontology , quantum mechanics , biology , programming language
The linearity of nine long Swedish macroeconomic time series, whose business cycle properties were discussed by Englund, Persson, and Svensson (1992), is tested and rejected for all but two. Non‐linear (STAR) models are estimated, and their properties are investigated. Business cycle frequency variation does not seem to be constant over time for all series; it is difficult to find a ‘Swedish business cycle’. Pairwise Granger non‐causality tests are adapted to the STAR case, and non‐causality is tested. The results point at strong temporal interactions and indicate that the functional form (linear or STAR) strongly affects the outcome of these tests. Copyright © 1999 John Wiley & Sons, Ltd.