Evaluating the Maximization-Maximization approach to measure default probabilities on structural credit risk models

• Autor(a) principal: Ventura, Cláudia Cristina Valério
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.14/25481
• topic: Domínio/Área Científica::Ciências Sociais::Economia e Gestão

This thesis implements the Maximization-Maximization (MM) algorithm proposed by Forte and Lovreta (2012), where in the first step the expected assets rate of return and the asset volatility are estimated applying the Maximum Likelihood technique. As the firm’s assets value is not observable, the observed equity values are treated as transformed data in order to derive the log-likelihood function. In a second step, the default barrier is estimated according to the interests of shareholders, corresponding to the optimal level considered for the firm to default, and as the one that maximizes their participation. Using a sample of fifty-five companies and a time period for the estimation of one year, our results prove that estimating the expected rate of return is hard and does not provide statistically significant results, as it is dependent and highly correlated to the observed equity values. The results for the five-year default probabilities computed were most of them equal to zero or too high.