Assessment of the Bankruptcy Risk in the Hotel Industry as a Condition of the COVID-19 Crisis Using Time-Delay Neural Networks

In this paper we demonstrate a new conceptual framework in the application of multilayer perceptron (MLP) artificial neural networks (ANNs) to bankruptcy risk prediction using different time-delay network (TDNN) models assess Altman’s EM Z″-score zones firms for sample 100 companies operating hotel industry Republic Serbia. Hence, accuracies 9580 forecasting ANNs trained period 2016 2021 are analyzed, and impact various input parameters ANN on their accuracy is investigated, including indicators, market internal nonfinancial lengths learning periods parameters, K-means clusters zones. Based research, 11 stability indicators (SIs) years under analysis formulated, which represent generalization capabilities models, i.e., differences errors between preceding year zone assessment given; these seen as consequence structural changes at level that occurred during relevant year. SIs validated through comparison with relative strength index (RSI) descriptive model, high correlation found. Special focus placed identification 2020 order COVID-19 crisis It established despite fact development Serbia highly volatile process, largest analyzed 2020, potential applications limited.