Abstract: Analysis of financial statement is one of the most important matters not only to corporate itself but also to intermediary organs, commercial banks and related government departments. In this paper, based on fuzziness of financial data and subjectivity of existing research methods, fuzzy multinomial time series model, fuzzy clustering and forecasting model are presented to financial statement sequence and comprehensive analyzing through samples choosing from listing corporations. Case study shows that these models are fit for solving these problems and the precision of testing are satisfying. So, it is practicable and reasonable to apply these methods to financial statement analyzing.