Abstract

The article considers the factors influencing the ratio of private and public investments in public-private partnership projects. Macroeconomic, organizational and financial characteristics of projects potentially affecting the share of the private partner in the project are highlighted. Using correlation and regression analysis, the direction and scale of this impact on the target feature is checked. On the basis of the World Bank database of PPP projects, the impact of the characteristics of PPP projects on the share of the private partner in their financing is estimated. As a result of the empirical study, the following characteristics have a significant impact: GDP deflator, national reserves, country risk premium, income level of the project implementation country, PPP project form, implementation industry, project term, as well as the total amount of investment in the project. On the basis of these characteristics, the classifiers of PPP projects by the share of a private partner are constructed using three algorithms: kNN, SVM, Decision Tree. As a result of testing the classifiers, the maximum accuracy of predictions was achieved by 74% using the Decision Tree algorithm.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call