Predicting market direction with hidden Markov models

This paper develops the model of Bicego, Grosso, and Otranto (2008) and applies Hidden Markov Models to predict market direction. The paper draws an analogy between financial markets and speech recognition, seeking inspiration from the latter to solve common issues in quantitative investing. Whereas previous works focus mostly on very complex modifications of the original hidden markov model algorithm, the current paper provides an innovative methodology by drawing inspiration from thoroughly tested, yet simple, speech recognition methodologies. By grouping returns into sequences, Hidden Markov Models can then predict market direction the same way they are used to identify phonemes in speech recognition. The model proves highly successful in identifying market direction but fails to consistently identify whether a trend is in place. All in all, the current paper seeks to bridge the gap between speech recognition and quantitative finance and, even though the model is not fully successful, several refinements are suggested and the room for improvement is significant.

Algorithms and methodologies for decision support in energy efficiency on buildings

Dissertação apresentada na faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para a obtenção do grau de Mestre em Engenharia Electrotécnica e de Computadores

Spatio-Temporal data modeling in response to deforestation monitoring (a case study of small region in Riau Province, Indonesia)

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Human activity recognition for an intelligent knee orthosis

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

Modelling HIV/AIDS length of stay in Portugal

Tese apresentada como requisito parcial para obtenção do grau de Doutor em Estatística e Gestão de Informação pelo Instituto Superior de Estatística e Gestão de Informação da Universidade Nova de Lisboa

Human activity recognition with accelerometry: novel time and frequency features

Human Activity Recognition systems require objective and reliable methods that can be used in the daily routine and must offer consistent results according with the performed activities. These systems are under development and offer objective and personalized support for several applications such as the healthcare area. This thesis aims to create a framework for human activities recognition based on accelerometry signals. Some new features and techniques inspired in the audio recognition methodology are introduced in this work, namely Log Scale Power Bandwidth and the Markov Models application. The Forward Feature Selection was adopted as the feature selection algorithm in order to improve the clustering performances and limit the computational demands. This method selects the most suitable set of features for activities recognition in accelerometry from a 423th dimensional feature vector. Several Machine Learning algorithms were applied to the used accelerometry databases – FCHA and PAMAP databases - and these showed promising results in activities recognition. The developed algorithm set constitutes a mighty contribution for the development of reliable evaluation methods of movement disorders for diagnosis and treatment applications.

Practice variations in the production of health

This study investigates three questions related to medical practice variation. First, it tests whether average length of stay across Portuguese National Health Service hospitals varies when controlling for differences in patients’ characteristics. Second, it looks at hospital-level characteristics in order to find out whether these are able to explain differences in average length of stay across hospitals. Finally, it proposes a best practice average length of stay for each of the six episodes of care analyzed. To perform the analysis, administrative data from the Diagnosis-Related groups’ data set for the year of 2012 was used. A replication of a hierarchical two-stage model with hospital fixed effects was carried out. The results show that after taking patients’ characteristics into account, variation in average length of stay across hospitals exists. This variation cannot be explained by hospital-level characteristics.

Enhancement of an implemented rating model and impact on existing risk management processes: A hands-on approach from financial supplier risk management within the automotive industry

The purpose of this paper is to conduct a methodical drawback analysis of a financial supplier risk management approach which is currently implemented in the automotive industry. Based on identified methodical flaws, the risk assessment model is further developed by introducing a malus system which incorporates hidden risks into the model and by revising the derivation of the most central risk measure in the current model. Both methodical changes lead to significant enhancements in terms of risk assessment accuracy, supplier identification and workload efficiency.

Finding the hidden influencers

This thesis is a first step in the search for the characteristics of funders, and the underlying motivation that drives them to participate in crowdfunding. The purpose of the study is to identify demographics and psychographics that influence a funder’s willingness to financially support a crowdfunding project (WFS). Crowdfunding, crowdsourcing and donation literature are combined to create a conceptual model in which age, gender, altruism and income, together with several control variables, are expected to have an influence on a funder’s WFS. Primary data collection was conducted using a survey, and a dataset of 175 potential crowdfunders was created. The data is analysed using a multiple regression and provided several interesting results. First of all, age and gender have a significant effect on WFS, males and young adults until the age of 30 have a higher intention to give money to crowdfunding projects. Second, altruism is significantly positively related to WFS, meaning that the funders do not just care about the potential rewards they could receive, but also about the benefits that they create for the entrepreneur and the people affected by the crowdfunding project. Third, the moderation effect of income was found to be insignificant in this model. It shows that income does not affect the strength of the relationship between the age, gender and altruism, and WFS. This study provides important theoretical contributions by, to the best of my knowledge, being the first study to quantitatively investigate the characteristics of funders and using the funder as the unit of analysis. Moreover, the study provides important insights for entrepreneurs who wish to target the crowd better in order to attract and retain more funders, thereby increasing the chance of success of their project.