Project Report – Machine Learning & Business Intelligence
Using Data Mining for Bank Direct Marketing
Project Presentation – Machine Learning & Anomaly Detection
Anomaly Detection in System-wide Data Collection
Master Thesis – Exposé
MT-Portfolio Optimization with Factor Views Proposal
Master Thesis – Research methods
Portfolio Optimization with Factor Views – Research Methods
Master Thesis – Abstract
In this thesis, I have implemented the Black-Litterman model with factor views as forward-looking forecasts constructed according to the average value and momentum, which can be measured in sector portfolios of the European equity market index STOXX 600. The goal was to investigate whether it is possible to use factor views, constructed from the past pricing information, in combination with the BL model, and if this approach results in improved risk-return properties of the optimized portfolios. Additionally, it was investigated if such optimized portfolios, and in which parameter setting, deliver risk-adjusted returns in excess of the STOXX 600 Index as a benchmark. The portfolio optimization was conducted on 10 sector portfolios defined by the STOXX Europe 600 Index universe in the period from 1999 to 2019. The factor portfolios were constructed using best 2 and worst 2 performing sectors according to the 12-week momentum and the book-to-market ratio respectively. First 5 years of data have been used for estimating the first sector covariance matrix and for computation of the factor views. The historical simulations have been performed from 2004 to 2019 using 4-week rebalancing period. My empirical findings show that, over the investigated period, the Momentum factor has shown higher premia relative to the Value factor. This fact has also been reflected in the resulting Black-Litterman optimized portfolios. The BL approach with the Momentum factor has resulted in superior risk-return portfolios relative to the benchmark. The optimization with the Value factor shows close to no positive effect on the portfolio characteristics. Surprisingly, using both factors in combination, yields no benefits over the Black-Litterman optimization with the Momentum factor. Contrary to the efficient market theory developed in the 1970s, by using the approach described in this research, under no transaction-costs condition and by using only publicly available data, I was able to outperform the European equity market in risk-adjusted terms by using a wide range of Black-Litterman framework parameter settings. However, the performance inevitably comes with additional factor risk, which must be regarded in further analysis. The presented BL factor approach is suitable for tilting diversified portfolios towards factors that are known to be performance relevant (Fama and French, 1992).
Bachelor Thesis – Abstract
The eHome project from the Vienna University of Technology  is an R&D project with goals of providing assistive technologies for private households of older people with idea to give them possibilities for longer and independent living in their homes. The eHome system consists of an adaptive intelligent network of wireless sensors for activity monitoring with a central context-aware embedded system . The primary goal of this thesis is to investigate unsupervised prediction and clustering possibilities of user behaviour based on collected timeseries data from infrared temperature sensors in the eHome environment. Three different prediction approaches are described. Hourly Based Event Binning approach is compared to two clustering algorithms, Hierarchical Clustering and Dirichlet Process GMM. Prediction rates are measured on data from three different test persons. This thesis first examines two different approaches for event detection from infrared signal data. In a second stage three different methods for unsupervised prediction analytics are discussed and tested on selected datasets. Clustering algorithms parameter settings for timeseries data have also been discussed and tested in detail. Finally the prediction performance results are compared and each method’s advantages and disadvantages have been discussed. The practical part of this thesis is implemented in IPython notebook. Python version was 2.7 on 64 bit Ubuntu linux 12.04 LTS. Data analysis has been implemented with Python’s Pandas library. Visualisations are made with Matplotlib and Seaborn libraries. The results reveal that prediction accuracy depends on data quantity and spread of data points. The simplest method in prediction comparison, the Hourly Based Binning has however given the best prediction rates overall. By contrast to the Hourly Based Binning the Dirichlet Process Gaussian Mixture Models clustering show best prediction performance on smaller training data sets and well spread data. By further parameter tuning on Dirichlet Process GMM clustering the prediction rates could be further improved coming very close or even over performing the Hourly Based Binning. Due to the unknown distribution and well spread data, choosing the right threshold parameter for the Hierarchical Clustering was trickier than initially assumed. Despite the initial assumptions for Hierarchical Clustering, this method was at least applicable for unsupervised prediction analytics on used data sets.