An important concept in the financial literature dealing with the modeling and forecasting of financial time series is that of weak efficiency. We apply a test of weak efficiency to a sample of securities of the Mexican Stock Market between 1991-2000. We use ARMA-GARCH modeling with series of different frequency and extrapolate from different time intervals to prove the exante predictive power of the models. We report results about those time series that generate the best forecast, as well as its average error and error interval.
