Modeling Customer Behavior of Non-Maturity Deposits

Abstract: The modeling of non-maturity deposits has become a highly relevant subject in the financial sector since these instruments constitute a significant portion of banks’ funding. A non-maturity deposit may look relatively simple, however, it has features that complicate the handling of these products. This thesis has the purpose of building a model based on the identification, integration and significance level of factors that influence customer behavior related to non-maturity deposits. Moreover, a mathematical approach based on a selection of these factors is made with the aim to analyze client behavior related to these products. The developed model uses simple linear regression and multiple linear regression with dummy variables to model long-term behavior. In contrast to the statistical methods that banks typically apply in this context, this thesis can contribute to the modeling of non-maturity deposits by highlighting customer behavior. Although, the evaluation of the mathematical approach indicates that the model might not be appropriate to use in real practice, it may arise ideas of alternative methods for the handling of non-maturity deposits.