Probabilistic Models in the Dynamics of Urban Projects Financing

This article explores the application of probabilistic models to analyze the expenditure dynamics of urban project financing. We employ Doob's decomposition for discrete submartingales as a mathematical framework. A general approach to modeling expenditure growth is described, based on representing the financing dynamics as a stochastic process consisting of a martingale and a compensator. Within the proposed model, the compensator reflects the planned expenditure growth, while the martingale represents random deviations from the plan. To validate the model's applicability, we analyzed project financing data from Moscow. The study identified trend components and random residuals, which were then tested for the martingale property using the Ljung-Box test. The analysis indicates a lack of statistically significant autocorrelation, supporting their interpretation as a martingale in Doob's decomposition. The findings suggest that the proposed method can be used for forecasting and analyzing the sustainability of budgetary financing, as well as for assessing the impact of random factors on the implementation of urban projects.

project financing, urban projects, submartingales, Doob decomposition