In this work we present a technique to obtain prediction intervals for chao
tic data. Using nearest neighbors method we give estimates of local varianc
e and percentiles of the prediction error distribution. This allows to defi
ne an interval containing a future value with a given probability. Its effe
ctiveness is shown with data generated by a chaotic economic model. (C) 200
1 Elsevier Science B.V. All rights reserved. JEL classification: C53; C22.