Computer Science · Artificial Intelligence
Do sunspot numbers cause global temperatures? Evidence from a frequency domain causality test
Rangan Gupta, Luis A. Gil-Alaña, OlaOluwa S. Yaya
See on the mapAbstract
This article applies the causality test in the frequency domain, developed \nby Breitung and Candelon (2006), to analyse whether sunspot numbers \n(used as a partial approximation to solar irradiance) cause global temperatures, \nusing monthly data covering the time period 1880:1–2013:9. \nWhile standard time domain Granger causality test fails to reject the null \nhypothesis that sunspot numbers do not cause global temperatures for \nboth full and sub-samples (identified based on tests of structural breaks), \nthe frequency domain causality test detects predictability for both the \nfull-sample and the last sub-sample at short (2–2.6 months) and long \n(10.3 months and above) cycle lengths, respectively. Our results highlight \nthe importance of analysing causality using the frequency domain \ntest, which, unlike the time domain Granger causality test, allows us to \ndecompose causality by different time horizons, and hence, could detect \npredictability at certain cycle lengths even when the time domain causality \ntest might fail to pick up any causality. Further, given the widespread \ndiscussion in the literature, those results for the full-sample \ncausality, irrespective of whether it is in time or frequency domains, \ncannot be relied upon when there are structural breaks present, and one \nneeds to draw inference regarding causality from the sub-samples, we \ncan conclude that there has been an emergence of causality running from \nsunspot numbers to global temperatures only recently at cycle length of \n10.3 months and above.