Estimating the amplitude of measurement noise present in chaotic time series

We propose a method for estimating the amplitude of measurement noise present in chaotic time series. This method is based on the evaluation of initial errors for a given time series and for a new one synthesized by adding an adequate amount of noise to the given one. The method is valid over a much wider range of noise levels than the previous methods are because it is not based on the detail of dynamical structure which generates the data. In addition, it is possible to check if the method is valid for the given data prior to its application. To confirm the effectiveness of the method we show the results of numerical experiments and apply the method to chaotic data obtained from an electrochemical experiment. (c) 1999 American Institute of Physics.