Climate change impacts are very dependent on regional geographic features and local climate variability. Impact assessment studies on climate change should therefore be performed at local or at most at the regional level for the evaluation of possible consequences. However, climate scenarios are produced by Global Circulation Models with spatial resolutions of several hundreds of kilometers. For this reason, downscaling methods are needed to bridge the gap between the large scale climate scenarios and the fine scale where local impacts happen. A stochastic weather generator, however, can serve as a computationally inexpensive tool to produce multiple-year climate change scenarios at the daily time scale which incorporate changes in both mean climate and in climate variability. In paper, LARS-WG model were used to downscale GCM outputs and then assessment of  the performance were done for generated daily data of precipitation, minimum and maximum temperature and sunshine hours. Study area is Ghare-su basin in Gorgan and the station is called Gorgan synoptic station. The first step is running the model for the 1970-1999 period. Then mean of observation and synthetic data were compared. T-test was used in the 95% significance level, and the difference between observation and synthetic data was not significant. Finally monthly mean of observation and synthetic data were compared using Statistical parameters such as NA, RMSE & MAE. As  a final result, it is found that performance of model is appropriate for generating daily above listed data in Ghare-su basin. So, it is possible to predict the climatic parameters from GCM output using LARS-WG model. Also minimum and maximum temperatures have highest and sunshine hours have lowest correlation.