Direkt zum InhaltDirekt zur SucheDirekt zur Navigation
▼ Zielgruppen ▼

Humboldt-Universität zu Berlin - Statistics

Prof. Dr. Brenda López Cabrera

JP Climate, Weather and Energy Analysis


[ Home | Teaching | Publications | Research Projects | CV | TalksLinks ]



Financial Statistics/Mathematics

  • Energy, Weather, Agricultural Markets
  • Quantitative Climate Finance
  • Carbon Finance
  • Insurance and Finance
  • Financial derivatives
  • Financial Risk Management
  • Portfolio optimization
  • Empirical and Computational Finance
  • Dimension reduction techniques
  • Time-varying parameter models and structural changes
  • Extreme value modelling


Former Students

  • Shi Chen - "Dynamic Adaptive Screening Hedging"
  • Franziska Schulz - "Weather and Energy Markets"
  • Thijs Benschop - "Carbon Finance"


  • "Application of Probabilistic Forecasting to Load Management" (2017 - ). Joint project with ÖKOTEC.

    The German energy transition (Energiewende) and the subsequent large increase in renewable energy production (from 6.2% in 2000 to 32.6% in 2015 (BMWi 2016)) has posed new challenges to participants of electricity markets. Besides the extension of the transmission grid and improvements in electricity storage, the flexibility of the demand side is a crucial part in facing these challenges. From a company perspective, participating in the market for flexibilities can also become a major opportunity. Though, this requires knowledge of future demand and ideally of the risk connected to it. This transfer project aims at the application of different forecasting methodologies, developed in the CRC 649 “Economic Risk”, to load management and the identification of flexibilities of industrial consumers. These flexibilities include controllable loads, peak shaving and adjustments in production. Together with ÖKOTEC Energiemanagement GmbH we want to test the applicability of the methods to load forecasting of industrial consumers and investigate the benefits for demand-side-management. Probabilistic forecasts (e.g. in form of interval forecasts) yield valuable information on the riskiness inherent in load scheduling and thus for risk management of industrial consumers. The contact of ÖKOTEC to leading players in industry yields the unique opportunity to obtain original data sets on electricity consumption at a high spatial resolution and in real time. This project aims at adapting the statistical models developed within the CRC to these new data sets. With the help of expert knowledge from ÖKOTEC, the goal is to develop a tool for load management and identification of flexibilities that is applicable to a wide range of data types. This tool can help both companies and policy makers. Companies benefit, since they can optimize their use of flexibilities. For policy makers, a lager degree of demand-side-flexibility will increase system stability and help to avoid extreme imbalances between power generation and consumption. This can reduce costs for storage capacities and grid infrastructure and result in more stable electricity prices.

  • Collective Cognition & Cooperation Network with Princeton University (2016 - ).

    Our interdisciplinary network provides a unique collaborative environment for research and exchange on the emergence of cooperation and collective cognition in human and animal systems. It includes scientists from such different fields as biology, physics, psychology, social sciences, economics, and engineering, with both empirical and theoretical backgrounds. Our aim is to focus on two main questions: (1) the impact of variable environments on evolution and persistence of cooperation, and (2) fundamental dynamics governing the spreading of behavior, so-called “behavioral contagion”, and its role in collective cognition. Specific attention will be paid to possible applications of the research towards promoting sustainability in coupled socio-economic-environmental systems and understanding collective risk perception in human and animal groups. A core part of the network activity will be (1) two interdisciplinary workshops, and (2) a research-based CoCCoN-Q-Course for senior undergraduates and young graduate students (Master & 1st year PhD level), which will give students the opportunity to gather early experience in international collaboration, while working on self-determined projects. A special methodological focus of the network will be novel visualization methods for presentation and analysis of complex network data.

  • DFG SFB 649 "Weather Risk Management" (Jan 2013 - 2016). Together with Martin Odening.

    Weather constitutes an important macroeconomic risk that affects a wide range of industries and the frequency and intensity of extreme weather events is expected even to increase. On the other hand, new markets have emerged on which weather risks can be exchanged and which support the development of risk management strategies. Against this background the overall objective of this project to assess the magnitude and the importance of weather related economic risks and to explore options to treat these risk with financial instruments. We focus on the agribusiness and the energy sector.

  • C.A.S.E. "Calibrating CAT bonds"

    After the occurrence of a natural disaster, the reconstruction can be financed with catastrophic bonds (CAT bonds) or reinsurance. For insurers, reinsurers and other corporations CAT bonds provide multi year protection without the credit risk present in reinsurance. For investors CAT bonds offer attractive returns and reduction of portfolio risk, since CAT bonds defaults are uncorrelated with defaults of other securities. As the study of natural catastrophe models plays an important role in the prevention and mitigation of disasters, the main motivation of this project is the pricing of CAT bonds for earthquakes and hurricanes under different trigger mechanisms. This project also focuses on their calibration of the bonds from different sides of the contract.

  • DFG - IRTG 1792 - High Dimensional Non Stationary Time Series. "Dynamic Factor models (DFM) for weather time series analysis" (2013 - 2016)

    The issue of global warming has received a great deal of attention recently. Due to its phenomenological nature, weather data is often high-dimensional, it contains nonlinear relationships between its variates and has long range dependencies. Dimension reduction may be done via Group Lasso techniques, but for such weather time series with heteroskedastic error structure it has never been analyzed before. By considering a more general loss function that describes the tail behavior of distributions from different angles, and bearing in mind the phenomenological nature of weather, we intend to implement DFM on weather risk related variables across different countries and measurements stations over time. Instead of estimating a plethora of parameters for each single individual curve, we will use DSFM to conditional expectiles and quantiles in combination Group Lasso techniques to reduce the number of estimated parameters. This is not only of particular importance for hedgers and speculators of weather risk Goodwin (2001) because of the precise prediction of tail events, but simply to better understand the characteristics of observed trends and confirm whether global warming is changing predictability of weather. The results will help us identify areas with similar weather patterns or areas of similar land usages and their embedded economic risks.

  • Assessing Wind Energy Potential, Pricing and Modelling (2014 - 2016) Together with 4initia GmbH, Ritter, M., Odening, M., Shen, Z. and Deckert, L.

    To meet the increasing global demand for renewable energy, such as wind energy, an increasing number of local wind conditions. Plain average wind speed maps cannot provide a precise forecast of wind power because of the non-linear relationship between wind speed and production. We suggest a novel, globally feasible approach to asess the loval wind energy potential.


Ladislaus von Bortkiewicz Chair of Statistics
JP Climate, Weather and Energy Analysis
School of  Business and Economics
Humboldt Universität zu Berlin
Unter den Linden 6
10099 Berlin, Germany