In software cost estimation, various models have been proposed for the prediction of the effort or the productivity of a software project. Although most of the traditional methods produce point estimates, in practice it is more realistic and useful for a method to provide interval predictions, i.e. ranges of values accompanied with probabilities that the true cost will fall in the interval. Intervals can also be used to assess the validity of a model or even the predictive power of the data. In this presentation, we review various models developed for classifying a new project to a cost category. Part of this work is supported by the ≥Telecommunications Software Process Optimization" ("DIERGASIA") project, which is co-funded by the European Social Fund and national resources (GSRT/PAVET-2005). The described methods will be applied to data from a company which develops telecommunication software and critical systems.