Connection of music and mathematics is known from Pythagoras. Music sounds are in harmony if it satisfies certain mathematical conditions. Despite tight connection between music and mathematics, the idea of using mathematical methods in musicology was used only in the 1950s. Last decade various methods of mathematical statistics, such as correlation analysis, time series analysis, analysis of variance, data clustering, Markov chain models was successfully applied in musiclogy. Though, there was no work done in this area in Lithuania. In this article musical data (notes) is analysed using statistical methods. Musical intervalas are analysed. Music compositions are classified. Analysis results are presented. Research uses works of a few Lithuanian composers.