The aim of this article is to present a forecast of budget revenue from the profit tax using econometric models. The set of applied models has to be reduced to very simple models due to short time series used. Therefore, the profit tax regression analysis is made in two stages. In the first stage, econometric modelling of profit tax revenue with the main profit indicators (called the profit tax base) is performed on the basis of information on profit tax regulation and its changes. In the second stage, algorithms of forecasting the profit tax base are formed when the main macroeconomic indicators of Lithuanian economy are used as regressors. Crossvalidation was applied to estimate the accuracy of these algorithms.