With climate change, data on throughfall, stemflow and rainfall interception in forests are much needed by hydrologists throughout the world, including the Mediterranean region. Such data are rare for Algeria, especially for cedar forests. Thus, the present study is interested in analysis of throughfall, stemflow and rainfall interception in a cedar forest dominated by Cedrus atlantica Manetti in the province of Batna (Eastern Algeria) and establishing models for forecasting the different rainfall fractions in this forest type. Specific containers (plastic bottles and collars) were installed at stems of 30 trees, under tree crowns and in an open area not covered by tree crowns between January 2018 and December 2019 to collect the different studied fractions of water (rainfall, throughfall and stemflow). Our results reveal that throughfall was with 65.41% of the total rainfall the greatest fraction in the studied cedar forest. Stemflow represented only a very small fraction of total rainfall with 3.86%. The mean rainfall interception was 30.72 % of the total rainfall. The relationships between rainfall, throughfall and stemflow were assessed by linear regressions. We found strong correlations between throughfall, stemflow and rainfall with coefficient of determination R2 ranging between 0.92 and 0.99. The equations developed by this study revealed that the studied cedar forest must receive at least 1.70 and 4.46 mm of rainfall, to trigger throughfall and stemflow, respectively.