This paper introduces neural networks technique for wind speed predict
ion and compares its performance with an autoregressive model. First,
we studied the statistical characteristics of mean monthly and daily w
ind speed in Jeddah, Saudi Arabia. The autocorrelation coefficients ar
e computed and the correlogram is found compatible with the real diurn
al variation of mean wind speed. The stochastic time series analysis i
s found to be suitable for the description of autoregressive model tha
t involves a time lag of one month for the mean monthly prediction and
one day for the mean daily wind speed prediction. The results on a te
sting data indicate that the neural network approach outperforms the A
R model as indicated by the prediction graph and by the root mean squa
re errors. (C) 1998 Elsevier Science Ltd. All rights reserved.