DIFFICULTIES IN USING AVERAGED ORBITS TO EXTRACT SCALING FUNCTIONS

Trajectory scaling functions are the basic element in the study of cha otic dynamical systems from which any long-time average can be compute d. They have never been extracted from an experimental time series, th e reason being their sensitivity to noise. It is shown, by numerical s imulations, that the scaling function is more sensitive to drift in th e control parameters than to noise in the time series. It is also expl ained how naive averaging of the orbit points may lead to erroneous re sults.