Continuation of the PhD grant due to obtaining the degree in advance. The focus of the current research lies primarily in combining results obtained as part of my PhD thesis with black-box machine learning techniques. By doing this, we believe that it is possible to further improve on reserving methods in non-life insurance.

How much capital must a non-life insurer put aside to stay solvent? This question has been the primary focus of my PhD research. My PhD thesis, titled "Micro-level claims reserving in non-life insurance", sets out to answer this question using statistical models, particularly with a focus on so-called micro-models. Micro-models are models that make full use of the extensive individual-level data that insurers nowadays, in the big-data era, have at their disposal. With this research, my ambition has been to bring the practising actuary closer to the rich mathematical theory that currently exists, in addition to contributing to its further development.The work during the PhD studies resulted in 5 papers published in peer-reviewed scientific journals.Teaching:20% of my PhD studies was committed to teaching, which included:- leading exercise sessions in front of the blackboard and supervising computer labs within basic probability theory, insurance mathematics, and survival analysis,- supervision of Bachelor theses, - review and examination of Bachelor and Master theses, and- supervision, review, and examination of larger course-specific projects within the course linear statistical models, which teaches the theory of linear regression models and ANOVA.Link to thesis:http://urn.kb.se/resolve?urn=urn:nbn:se:su:diva-174300

The focus of the thesis lies in examining the performance of some survival analysis methods concerning extrapolation of survival curves. Extrapolating well is essential when we have data of limited follow-up available. To do this, a real-world registry dataset consisting of Swedish multiple myeloma patients between the years 2000 and 2011 forms the foundation of a simulation study which aims to assess the performance of the different survival analysis methods.Link to thesis:http://kurser.math.su.se/pluginfile.php/20130/mod_folder/content/0/Master/2016/2016_02_report.pdf

Consulting at the Department of Neuroscience.The project which formed the main part of my employment involved using statistical methods to help neuroscientists map the connections between neurons in the brains of mice. In addition to this, I functioned as a consulting statistician to the researchers at the lab.