In several climate change impact studies, outputs (especially rainfall) of global and regional climate model are important inputs for hydrological and hydraulic models. Therefore, the quality of precipitation data needs to be assessed. The study conducted to compare downscaled PRECIS precipitation with observed precipitation and to apply a bias correction to this variable. The bias correction has practiced by fitting downscaled data to observed data in terms of frequency and intensity. In comparison with observed data, frequency of downscaled precipitation is significantly greater while intensity is considerable lower than the observations. The bias correction is calculated using data of a decade and validated using another decade before applying for predicted data. Results show that the method performs well; not only precipitation frequency is improved but also precipitation intensity is fitted to the observations. This results encourages the application of the method to multi-decadal precipitation projections.