Determination of precipitation data, which is an important parameter in meteorological conditions; it directly affects many sectors such as agriculture, energy, environment, transportation, logistics and disaster control. Correct prediction of future precipitation data will make the daily life of people easier, especially. Using rainfall data from previous periods will allow precipitation forecasts for future periods and establishment of early warning systems. In this study, Turkey's Central Anatolia Region stations rainfall predictions for the future are made using monthly total precipitation data from 1990−2015 year. Two different approaches have been taken for precipitation predictions. The first of these approaches is to apply only artificial neural network (ANN) and adaptive neural fuzzy inference system (ANFIS) to monthly total precipitation data of stations belonging to previous periods. Another approach is to apply the wavelet transform model to both ANN and ANFIS. In these models, monthly total precipitation data between 1990 and 2009 were used as training data, and data between 2010 and 2015 were used as test data. Precipitation data were used as input data for times t−2 and t−1, and precipitation at time t was tried to be predicted. The prediction performances of these models were evaluated according to statistical criteria such as determination coefficient. According to the coefficient of determination (0.011−0.189), the predictions made by applying ANN and ANFIS to the precipitation data give bad results. According to the coefficient of determination, ANN with wavelet transform (WT−ANN) and ANFIS with wavelet transform (WT−ANFIS) models are more successful in predicting future precipitation data. In addition, one of the important results of this study according to statistical criteria is that the WT−ANN and WT−ANFIS models give very similar results to each other.