Disputas: Noor ’Adilah Ibrahim

M.Sc. Noor ’Adilah Ibrahim ved Matematisk institutt vil forsvare sin avhandling for graden ph.d.:

Stochastic Modelling of Weather Risk in Energy Markets

Noor ’Adilah Ibrahim

Tid og sted for prøveforelesning

17. januar 2019 kl. 10.15, "Abels utsikt", 12. etasje, Niels Henrik Abels hus.

Bedømmelseskomité

  • Professor Silvana Stefani, Universitá Milano-Bicocca

  • Professor Delphine Lautier, Universite Paris-Dauphine

  • Associate Professor Tenure track Kristina Rognlien Dahl, Universitet i Oslo

Leder av disputas

Instituttleder Geir Dahl, Matematisk institutt, Universitet i Oslo

Veiledere

  • Professor Fred Espen Benth, Matematisk institutt, Universitet i Oslo

  • Førsteamanuensis Ingrid Hobæk Haff, Matematisk institutt, Universitet i Oslo

  • Professor Almut Veraart, Imperial College London

Sammendrag

In Europe, photovoltaic (PV) and wind energy are becoming major carbon-intensive power production substitutes. Germany, for instance, is undergoing Energiewendewhere they utilize renewable power to reduce carbon emission. However, PV and wind are weather dependent and lead to intermittent productions. The uncertainty in the productions may cause troubles to many parties especially to the energy producers.

The objectives of the thesis are to answer some of the unanswered research problems concerning the best prediction model which can capture both deterministic (long-term) and stochastic (short-term) factors of interday and intraday behavior, as well as the dependency between PV and wind energy productions. Our main finding shows that the sun intensity function is very helpful in explaining the long-term behavior of the production, while the autoregressive process (AR) with order two is sufficient enough to explain its short-term dynamics. We are also able to capture the effects of sunrise and sunset, but a model refinement is required to explain the seasonality of the intraday productions. In addition, we observe a negative correlation between PV production and electricity prices, but there is no clear dependency detected between residuals of PV and wind. As a hedging strategy against volume and price risks, we construct quanto options which might be beneficial to the non-renewable energy producers. 

We also perform a case study on temperature futures using polynomials processes. Undeniably, the weather risks are mostly induced by the temperature changes. Until recently, many industries use weather derivatives, where the most liquid is based on the temperature. Therefore, it is highly necessary to formulate the price of the temperature derivatives using an appropriate approach. We observe that the polynomials approach does not necessarily bring in any benefits in pricing compared to the introduced approach such as a distributional method.

 

For mer informasjon

Kontakt Matematisk institutt

Publisert 13. des. 2018 16:02 - Sist endret 17. jan. 2019 09:00