Abstract
A statistical approach for chaos identification in a time series is described and applied to numerical data generated from Chua's circuit. This method compares the short-term predictability for a given time series to an ensemble of random data which has the same Fourier spectrum as the original time series. The forecasting error is computed as a statistic for performing statistical hypothesis testing. The forcasting technique is modified by introducing a moving predictor. The results show that this will give more accurate predictions, hence, better capability of distinguishing chaos from random noise in time series.