Principal component analysis of the yield curve

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Finance from the NOVA – School of Business and Economics

Three essays on the term structure of interest rates

A PhD Dissertation, presented as part of the requirements for the Degree of Doctor of Philosophy from the NOVA - School of Business and Economics

Credit default swap (CDS) prediction model & trading strategy

This project focuses on the study of different explanatory models for the behavior of CDS security, such as Fixed-Effect Model, GLS Random-Effect Model, Pooled OLS and Quantile Regression Model. After determining the best fitness model, trading strategies with long and short positions in CDS have been developed. Due to some specifications of CDS, I conclude that the quantile regression is the most efficient model to estimate the data. The P&L and Sharpe Ratio of the strategy are analyzed using a backtesting analogy, where I conclude that, mainly for non-financial companies, the model allows traders to take advantage of and profit from arbitrages.

Broadband: What do consumers want? Examining willingness-to-pay

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Economics from the NOVA – School of Business and Economics

Returns to vocational education in Portugal

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Economics from the NOVA – School of Business and Economics

Previsão da estrutura temporal da taxa de juro na zona Euro : aproximação paramétrica e por métodos de aprendizagem automática

A estrutura temporal das taxas de juro, também conhecida por yield curve ou curva de rendimentos define a relação entre as taxas de juros e o prazo de vencimento (ou maturidades) dos investimentos feitos. Assim, o desenvolvimento de modelos que possibilitem a obtenção de previsões precisas sobre a estrutura temporal das taxas de juro e que permitam estudar a dinâmica da evolução das taxas de juro é de crucial importância em diversas áreas de financiamento. Neste estudo investigou-se a performance de diferentes métodos de previsão para obter a estrutura temporal das taxas de juro da Zona Euro, considerando o período entre 2009 e 2015. Em termos mais específicos, foi analisada a capacidade preditiva do modelo de Nelson-Siegel & Svensson assumindo que os parâmetros resultantes da estimação da especificação paramétrica podem ser modelizados através de métodos de séries temporais univariados (modelos ARIMA, Random walk) e multivariados (modelos VAR) e Redes Neuronais Artificiais (RNA) individuais e conjuntas. Os resultados deste estudo mostram que (i) as RNA com a previsão dos parâmetros em simultâneo exibem os valores de erro mais baixos para as maturidades de curto e médio prazo (3 meses a 5 anos); (ii) As RNAs individuais são melhores para prever as taxas de juro nas maturidades compreendidas entre os 7 e os 10 anos, e que (iii) para as maturidades de longo e muito longo prazo (15 e 30 anos respetivamente) deverá ser escolhido o modelo VAR(1). Estes resultados são robustos e consistentes para todos os horizontes de previsão analisados (1,2 e 3 meses). Contudo, no período analisado nenhum dos modelos testados apresenta valores de erro inferiores aos obtidos com o modelo Random Walk.

Sonae MC: The road model of Kaizen

Sonae MC is considered the first success case of Kaizen in the retail industry. Before becoming a true role model for so many companies, there was a long road to walk. However, it may still be hard to understand the steps taken on the way. How could a training program develop into an integral continuous improvement system, and how did it affect the company – its people, culture, operations and strategy? How was it possible to get everyone on board? How could it be sustained until today, when Kaizen usually fails in the West? What were the critical factors for success?