Climatic data are extremely important in many human activities; they provide much information about the environment and its impacts. So there is a need for weather data information, statistically homogeneous, because heterogeneity can compromise any time series analysis and interpretations of such data. Thus, this work aimed to study methodologies for assessing the homogeneity of data sets of maximum and minimum temperatures in the state of Minas Gerais, Brazil. The analysis of homogeneity was carried out by means of three tests for the identification of points of discontinuity in the series: Standard Normal Homogeneity Test (SNHT), Pettitt test and the Buishand test. The results of the tests were very similar, with most of the monthly series of temperature were considered homogeneous. Heterogeneities occurred in the 1990s, especially around 1997.
time series; Pettitt test; Buishand test
