In this study, regional and point frequency analysis of total annual precipitation were carried out for 1975-2010 period over Central Anatolia Region in Turkey. Precipitation data have been obtained from 47 raingauges. In the first case, when whole raingauges were assumed as one region, there were some discordant sites according to diagnostic tests namely discordancy and heterogeneity measures. When Central Anatolia Region was split up four region according to cluster analysis using site characteristic vectors, there were no discordant sites in that regions. So, it was decided that the data having stations in the climatologically homogeneous regions were useful for regional and point estimation. The best-suitable distributions have been selected as generalized normal for region one, generalized extreme values for region two and three and Pearson type 3 for region four through goodness-of-fit test for climatologically homogeneous regions. Precipitation estimation were determined for some return periods using regional L-moment algorithm. In the second case, Monte Carlo simulation approach was used to evaluate the robustness of the regional and point estimation computing some accuracy measures.