Bankruptcy Prediction for Micro and Small Enterprises Using Financial, Non-Financial, Business Sector and Macroeconomic Variables: The Case of the Lithuanian Construction Sector

Credit-risk models that are designed for general application across sectors may not be suitable the construction industry, which has unique characteristics and financial risks require specialised modelling approaches. Moreover, advanced bankruptcy-prediction often used to achieve highest accuracy in large modern datasets. Therefore, aim of this research is creation enterprise-bankruptcy prediction (EBP) Lithuanian micro small enterprises (MiSEs) sector. This issue analysed based on classification specific types variable used. Firstly, four proposed. In EBP models, variables substantially explain an enterprise’s statements performance from different perspectives. Including enterprises’ non-financial, construction-sector macroeconomic improves models. The inclusion model a particularly significant impact. These findings can great significance investors, creditors, policymakers practitioners assessing making informed decisions. second question related To develop logistic regression (LR), artificial neural networks (ANNs) multivariate adaptive splines (MARS) were addition, study developed two-stage hybrid i.e., LR combined with ANNs. show do improve bankruptcy prediction. It cannot argued ANN more accurate predicting bankruptcy. MARS demonstrates best prediction, could valuable tool stakeholders evaluate risk.