Functional data analysis for optimizing strategies of cash flow management

The cash management deals with problem of automating and managingcash flow processes. Optimization of the management processes greatly reducesoverall cash handling costs. The present analysis is an empirical study of cash flows,from and to bank branches, deriving an underlying theoretical framework, which canin a reasonable way be connected with the optimal strategy. Functional data analysisis considered an appropriate framework to analyse the dynamics of the time seriesbehavior of cash flows: since the observations are not equally spaced in time andtheir number is different for each series, they are converted in a collection of randomcurves in a space spanned by finite dimensional functional bases. A central issue inthe analysis is describing specific patterns of the curves, taking into account thetemporal dependence, and the dependence between curves. The analysis providesa dynamic cash management model that is applied with alternative strategies forprogramming a cash in transit for the difference between cash inflows and cashoutflows in a fixed interval of time. As the strategies are affected by changes inthe behavior of the cash flows, the dynamic model outperforms more traditionalapproaches in identifying the optimal strategy.