Time series modelling of sunspot numbers using long range cyclical dependence

This paper deals with the analysis of the monthly structure of sunspot numbers using a new technique based on cyclical long range dependence. The results show that sunspot numbers have a periodicity of 130 months, but more importantly, that the series is highly persistent, with an order of cyclical fractional integration slightly above 0.30. That means that the series displays long memory, with a large degree of dependence between the observations that tends to disappear very slowly in time