The University of Oslo is closed. The PhD defence and trial lecture will therefore be digital and streamed directly using Zoom. The host of the session will moderate the technicalities while the chair of the defence will moderate the disputation.
Ex auditorio questions: the chair of the defence will invite the audience to ask questions ex auditorio at the end of the defence. If you would like to ask a question, click 'Raise hand' and wait to be unmuted.
-
Join the disputation
The webinar opens for participation just before the disputation starts, participants who join early will be put in a waiting room.-
Submit the request to get access to the thesis (available from 4th June 13:15, until 18th June 13:15)
Trial lecture
18th June, kl.10:15, Zoom
"Machine Learning methods for commodity markets trading and forecasting"
-
Join the trial lecture
The webinar opens for participation just before the trial lecture starts, participants who join early will be put in a waiting room.
Main research findings
The production of renewable energy is growing world-wide, and -- as a result -- power production is becoming increasingly dependent on weather factors such as temperature, wind and precipitation. All of these factors are hard to predict, and this causes power prices to change rapidly and unpredictably, and makes the modelling of financial risk in energy markets particularly challenging. This thesis develops new models and tools to be used in this direction.
Energy markets can be divided into three main sectors: there is (1) a spot market for short-term delivery contracts, (2) a forward market for delivery in a future time at a price set today, and (3) an option market where the contracts traded allow, but not oblige, the buyer to buy or sell the asset in a future time at a price set today. Buying and selling electricity in these markets, while managing the financial risk, requires accurate mathematical models.
This thesis is concerned with the modelling of these markets. It both develops concrete models and more abstract mathematical tools which can be used for this challenging task. In particular, it focuses on spot price modelling, by taking into account the dependence between spot price behaviour and weather variables such as wind speed. Moreover, it focuses on forward price modelling and pricing of options written on forward contracts with delivery period, which are typical in the energy markets. Two central challenges which are addressed in this thesis are model accuracy and computational complexity, both of which are improved upon by using deep learning.