نوع مقاله : مقاله پژوهشی
نویسندگان
1 دانشیار مدیریت مالی دانشگاه تهران
2 عضو هیات علمی دانشگاه صنعتی شاهرود
3 عضو هیات علمی دانشگاه تهران
4 دانشجوی دکتری پردیس البرز دانشگاه تهران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Most managers, shareholders, creditors, and other stakeholders are looking for the practical tools to evaluate and predict companies' financial health. Having such models give them enough time to make right decisions before falling into the bankruptcy.
In the following research, a new early warning model for prediction of companies' financial health using hybrid Bat Artificial neural network developed.
The research population encompasses Iranian public manufacturing companies, and sample used in the model includes 80 random healthy and bankrupt manufacturing companies listed in the Tehran stock market from 2008 to 2017. Also, for developing the model, 12 various and independent financial ratios as explanatory variables employed.
For performance evaluation, the outputs of suggested hybrid BAT-ANN model compared to the Altman model and back-propagation artificial neural networks. The proposed model is able to predict signs of companies' financial unhealthiness up to 3 years before the failure. The statistical evidence demonstrated that the hybrid model outperformed the other two models at 5% significant level.
کلیدواژهها [English]