We discuss application of F-transform with respect to raised cosine
generalized uniform fuzzy partition to analysis of time series
under assumption that the time series can be additively decomposed
into trend-cycle, seasonal component and noise. We focus that when
setting reasonably width of the basic functions, inverse
F-transform of the time series is good approximation of its
trend-cycle. This mean that we can determine a generalized fuzzy
partition such that F-transform completely remove seasonal
component and noise.