Previsão de rating empresarial com o uso de índices contábeis
Credit rating is a tool that provides interested professionals with an efficient way to analyze a company’s credit risk. In this work, forecast models for the credit rating of Brazilian companies were estimated, using the S&P credit rating and accounting ratios for the period 2017 to 2019. The r...
| Autor: | |
|---|---|
| Formato: | tesis de maestría |
| Estado: | Versión publicada |
| Fecha de publicación: | 2021 |
| País: | Brasil |
| Recursos: | Universidade Federal do Ceará (UFC) |
| Repositorio: | Repositório Institucional da Universidade Federal do Ceará (UFC) |
| Idioma: | portugués |
| OAI Identifier: | oai:repositorio.ufc.br:riufc/59886 |
| Acesso em linha: | http://www.repositorio.ufc.br/handle/riufc/59886 |
| Access Level: | acceso abierto |
| Palavra-chave: | Rating de crédito Probit ordenado Lasso Contabilidade |
| Resumo: | Credit rating is a tool that provides interested professionals with an efficient way to analyze a company’s credit risk. In this work, forecast models for the credit rating of Brazilian companies were estimated, using the S&P credit rating and accounting ratios for the period 2017 to 2019. The rating of the companies in the sample is between B and AAA ratings. The first estimated model was the ordered post-lasso probit with five statistically significant variables. The results of this model corroborate those obtained by Damasceno et al. (2008), but they have a higher hit rate, correctly predicting 72.39% of the sample. The model had a low hit rate for rating categories B, A and AAA, while category AA had a high hit rate of 97.65%. The third estimated model was the ordered probit model with the variables selected by Damasceno et al. (2008), using data from this survey. The two variables present in the model were statistically significant. The hit rate of the model was 64.92%, being lower than the first model. The third model was not able to predict any ratings B or AAA, but correctly predicted all ratings AA. |
|---|