How Political and Economic Freedom Affect Modelling of FDI Determinants? A Local Linear Neuro-Fuzzy Approach

This paper presents a novel approach to model determinants of - foreign direct investment (FDI) and the influence of institutional factors to improve the model. The proposed approach includes two main steps. First, a local linear neuro-fuzzy (LLNF) model is proposed to model the FDI determinants. The LLNF model is composed of a number of local linear models and trained by a heuristic fast and efficient learning algorithm. In the next step, the available dataset of 118 countries for the period of 2000 to 2019 is classified into smaller subsets based on the level of political freedom (three subsets) and another time based on economic freedom index (four subsets). Then, for each subset, the FDI is modeled using LLNF approach. This step allows us to find out whether economic freedom or political freedom has an influence on FDI or not without having those variables in the model. In other words, based on the two-step approach we are able to thoroughly investigate how strongly the FDI is affected by the economic freedom and political freedom. The results of this paper’s examination highlight that economic freedom is important for determining FDI, but such conclusion cannot be made about political freedom