Detalhes bibliográficos
| Ano de defesa: |
2017 |
| Autor(a) principal: |
Pires, César Augusto
 |
| Orientador(a): |
Oliveira, Antonio Benedito Silva |
| Banca de defesa: |
Não Informado pela instituição |
| Tipo de documento: |
Dissertação
|
| Tipo de acesso: |
Acesso aberto |
| Idioma: |
por |
| Instituição de defesa: |
Pontifícia Universidade Católica de São Paulo
|
| Programa de Pós-Graduação: |
Programa de Estudos Pós-Graduados em Ciências Contábeis e Atuariais
|
| Departamento: |
Faculdade de Economia, Administração, Contábeis e Atuariais
|
| País: |
Brasil
|
| Palavras-chave em Inglês: |
|
| Área do conhecimento CNPq: |
|
| Link de acesso: |
https://tede2.pucsp.br/handle/handle/20320
|
Resumo: |
This paper aims to identify the accounting performance indicators through techniques applied in companies that signal the judicial recovery using logistic regression and discriminant analysis, according to its relevance because it seeks to help decision making by the corporate body of Organizations to avoid future financial problems. The origin and evolution of bankruptcy legislation in Brazil and several models of insolvency used in the literature were presented in research, because it is a descriptive research in relation to its objectives, and quantitative, in terms of procedures, using statistical analysis techniques to evaluate the performance of classification techniques applied to the insolvency problem of publicly held companies; documents and accounting data from 2005 to 2015 were collected from the BM & FBovespa database for the application of empirical tests. The discriminant analysis was able to e valuate 88% of the cases correctly, which is a good percentage of prediction and does not present type II error, that is, to classify a solvent company in judicial recovery, and with 11 variables, since one was discarded, but when logistic regression is compared to discriminant analysis, it provides predictive accuracy comparable to a simpler statistical variable that used the same substantial interpretation with only one variable less and with a global 90% hit percentage. From the results of the logistic regression, it is possible to focus only on the variables X4 = asset structure and X2 = Return concerning equity as the main ones in the differentiation of groups, since the goal of the analysis is not to increase the likelihood of success, once that logistic regression provides a direct technique to distinguish firms' judicial recovery from solvent enterprises and to understand the relative impact of each independent variable in creating differences between the two gro ups of firms. Finally, the results presented show that logistic regression, even using a smaller number of variables, holds a better percentage of correctness |
