Cash flow forecasting using Continuous-Time Stochastic Processes

Since the liquidity situation is the basis for many people to judge the position of the economic unit, this issue has been considered by stakeholders including creditors and investors.The purpose of this study is a new understanding of cash flow behavior and cash balance forecasting. The statistical population of the present study is the annual cash balance of 48 branches of a certain bank. For this purpose, the optimal model out of 4 models; Geometric Brownie, Arithmetic Brownie, Vasicek and Modified Square Root Model at three levels of microscopic, mesoscopic and macroscopic have been investigated and the geometric Brownie model has been approved as the optimal model; Then, using the mentioned optimal model, the cash balance is predicted in different time horizons. The result shows that the strength of the GBM model does not remain constant with increasing the length of the forecast time horizon and with increasing the forecast time horizon, they have different results. The accuracy of the model forecast with the MAPE criterion at the 14-day horizon is at an appropriate level.