A. Dimitriadis et Am. Fraser, MODELING DOUBLE SCROLL TIME-SERIES, IEEE transactions on circuits and systems. 2, Analog and digital signal processing, 40(10), 1993, pp. 683-687
The ubiquity of strange attractors in nature suggests that nonlinear m
odeling techniques can improve performance in some signal processing a
pplications. We introduce mixed state Markov models (MSMM's), a refine
ment of hidden filter HMM's, and apply both to a synthetic double scro
ll time series. Forecasts by HFHMM's diverge after a few steps. Using
ad hoc procedures, forecasts by MSMM's, even models generated by crude
methods without iterative optimization, can be made more stable.