Developing a model for predicting bankruptcy of construction industry enterprises

Abstract

Subject. The article considers methods of forecasting bankruptcy of a construction industry enterprise. Objectives. The study aims to develop an industry model to forecast bankruptcy of construction companies. Methods. The study employs general scientific methods of cognition, like analysis, synthesis, comparison, generalization, observation. The main statistical and economic research method is linear multidimensional discriminant analysis (MDA), which is used to perform calculations in the program for statistical data processing, SPSS Statistics. Results. We selected optimal predictors for each group of indicators of profitability, liquidity, financial stability, and business activity, which can explain company’s financial health as a whole, based on financial information. Furthermore, we developed and tested an industry model to predict bankruptcy of construction enterprises. Its predictive accuracy is 81.7%, indicating the effectiveness of the model. Conclusions. The developed model, created using an industry sample, provides a more accurate assessment of bankruptcy probability of construction industry enterprises compared to "universal" models. Using this model as an auxiliary tool for analyzing the financial condition of enterprises is an effective method of detecting and timely preventing bankruptcy. When changing weight coefficients, this model can also be applied in other economic sectors.