Investigating long range dependence in temperatures in Siberia

In this paper we examine monthly mean temperatures in 40 selected stations Siberia for the time period January 1937–December 2020 using long range dependence techniques. particular, use a fractionally integrated model that incorporates linear trend along with seasonal structure. Our results show first memory is present all significantly positive values differencing parameter, though, at same component seems to be important cases. Performing unit root tests, support nonstationary seasonality and working differenced data, differ depending on structure of error term: if errors are uncorrelated, present; however, allowing autocorrelation, feature disappears favor short pattern.