An improved drought management must rely on an accurate monitoring and forecasting of the phenomenon in order to activate appropriate mitigation measures, based on the risk associated with the possible evolution of a current drought condition. In the paper, a stochastic model to forecast SPI values at short-medium term has been developed. The forecast is based on the expected value of future values of SPI, conditioned on past observations of precipitation and the resulting analytical expression is a function of past precipitation and of its statistics. Also, analytical expressions of the Mean Square Error (MSE) and of prediction confidence limits of fixed probability are derived, which enable to assess the forecast accuracy. Validation of the model has been carried out with reference to precipitation series observed in 43 stations located in Sicily, Italy, making use of a moving window scheme for parameters estimation. Results show a good agreement between observed and forecasted values, thus suggesting the suitability of the proposed procedure as a tool for drought management.