Based on downscaled daily precipitation values from the global climate model of the Max Planck Institute in Hamburg, time series of 20 years have been generated to describe the current climate of 1980–1999 (control data) and the future climate of 2030–2049 (scenario data) for Norway. These time series serve as training data for the Randomised Bartlett-Lewis Rectangular Pulse Model (RBLRPM), a precipitation simulation model, and time series of 1000 years length have been generated to assess possible changes in the extreme precipitation regime due to climate change. The analysis of changes in extreme value patterns for annual and seasonal values in the scenario and control data sets shows tendencies towards increased extreme values and seasonal shifts for the scenario period. A general increase in mean and standard deviation of the extreme value sample and for values of 10 and 100 years return period is found, although the regional variability is significant. For some regions the increase is in the order of 10 to 50 % for both annual and seasonal values.