Modelo h??brido de avalia????o de risco de cr??dito para corpora????es brasileiras com base em algoritmos de aprendizado de m??quina

Submitted by Sara Ribeiro (sara.ribeiro@ucb.br) on 2018-08-08T13:33:03Z No. of bitstreams: 1 RafaelLeiteGregorioDissertacao2018.pdf: 1382550 bytes, checksum: 9c6e4f1d3c561482546aca581262b92b (MD5) Approved for entry into archive by Sara Ribeiro (sara.ribeiro@ucb.br) on 2018-08-08T13:33:24Z (GMT) No. of bitstreams: 1 RafaelLeiteGregorioDissertacao2018.pdf: 1382550 bytes, checksum: 9c6e4f1d3c561482546aca581262b92b (MD5) Made available in DSpace on 2018-08-08T13:33:24Z (GMT). No. of bitstreams: 1 RafaelLeiteGregorioDissertacao2018.pdf: 1382550 bytes, checksum: 9c6e4f1d3c561482546aca581262b92b (MD5) Previous issue date: 2018-07-09 The credit risk assessment has a relevant role for financial institutions because it is associated with possible losses and has a large impact on the balance sheets. Although there are several researches on applications of machine learning and finance models, a study is still lacking that integrates available knowledge about credit risk assessment. This paper aims at specifying the machine learning model of the probability of default of publicly traded companies present in the Bovespa Index (corporations) and, based on the estimations of the model, to obtain risk assessment metrics based on risk letters. We converged methodologies verified in the literature and we estimated models that comprise fundamentalist (balance sheet) and governance data, macroeconomic and even variables resulting from the application of the proprietary model of KMV credit risk assessment. We test the XGboost and LinearSVM algorithms, which have very different characteristics among them, but are potentially useful to the problem. Parameter Grids were performed to identify the most representative variables and to specify the best performing model. The model selected was XGboost, and performance was very similar to the results obtained for the North American stock market in analogous research. The estimated credit ratings suggest that they are more sensitive to the economic and financial situation of the companies than that verified by traditional Rating Agencies. A avalia????o do risco de cr??dito tem papel relevante para as institui????es financeiras por estar associada a poss??veis perdas que podem gerar grande impacto nos balan??os. Embora existam v??rias pesquisas sobre aplica????es de modelos de aprendizado de m??quina e finan??as, ainda n??o h?? estudo que integre o conhecimento dispon??vel sobre avalia????o de risco de cr??dito. Este trabalho visa especificar modelo de aprendizado de m??quina da probabilidade de descumprimento de empresas de capital aberto presentes no ??ndice Bovespa (corpora????es) e, fruto das estima????es do modelo, obter m??trica de avalia????o de risco baseada em letras (ratings) de risco. Convergiu-se metodologias verificadas na literatura e estimou-se modelos que compreendem componentes fundamentalistas (de balan??o) e de governan??a corporativa, macroecon??micos e ainda vari??veis produto da aplica????o do modelo propriet??rio de avalia????o de risco de cr??dito KMV. Testou-se os algoritmos XGboost e LinearSVM, os quais possuem caracter??sticas bastante distintas entre si, mas s??o potencialmente ??teis ao problema exposto. Foram realizados Grids de par??metros para identifica????o das vari??veis mais representativas e para a especifica????o do modelo com melhor desempenho. O modelo selecionado foi o XGboost, tendo sido observado desempenho bastante semelhante aos resultados obtidos para o mercado de a????es norte-americano em pesquisa an??loga. Os ratings de cr??dito estimados mostram-se mais sens??veis ?? situa????o econ??mico-financeira das empresas ante o verificado por ag??ncias de rating tradicionais.