The objective of this paper is to identify best fit probability distribution of eight station of Marathwada region. Based on this fitting probability distribution annual rainfall value of different return period estimate. Marathwada region consists of eight districts. The total area of the region is 64,798 sq. km. The annual rainfall data 32 yea (1978 to 2010) have been collected IMD. Three statistical goodness of fit test such as Kolmogorov–Smirnov test (K-S), Anderson Darling test (A²) and Chi-Square test (χ²) was applied to annual rainfall data to select best fit probability distribution among 17 probability distribution. The select best fit probability distribution on the basis of highest rank with minimum value of test statistic. Fourth probability distribution was identified using maximum overall score based on sum of individual point score obtained from three selected goodness of fit test. On the basis of fitting probability distribution estimate parameter and return period of annual rainfall. The analysis of data indicated that different distribution fitted. Aurangabad district it was Weibull, For Jalna it was Weibull (3P), Latur it was Lognormal (3P), for Nanded it was Gamma Distribution, for Osmanabad it was Weibull and for Parbhani district fitted distribution was Gamma (3P).