Workshop #3. – Statistical Inference in Energy Markets

Institut Henri Poincaré, Paris - May 25 2018

Confirmed speakers: Mark Podolskij (Aarhus University), Paulina Rowinska (Imperial College, London), Alexandre Brouste (Le Mans Université), Markus Bibinger (Philipps Universität, Marburg), John Moriarty (Queen Mary University of London), Rafał Weron (Politechnika Wrocławska)



9:00 - 9:30: Welcoming of the participants

09:30 – 10:15 Alexandre Brouste (Le Mans Université), Parametric estimation at high-frequency


Asymptotic efficiency of the sequence of maximum likelihood estimators is considered in statistical experiments implying the fractional Gaussian noise or symmetric stable random variables observed at high-frequency. Likelihood ratio hypothesis tests are also studied with an application to oil price modeling.

10:15 – 11:00 Mark Podolskij (Aarhus University), Statistical inference for fractional models


In recent literature fractional and moving average type models have gained popularity in economics and finance. Examples include fractional Brownian/stable motion, rough volatility models and Hawkes processes. In this talk we will review some existing estimation methods and present new results. We will make the link to potential application in modelling energy markets


11:00 - 11:30 Coffee break

11:30 - 12:15 Paulina Rowinska (Imperial College), Blowing in the Wind


We introduce a three-factor model of electricity spot prices, consisting of a deterministic seasonality and trend function as well as short- and long-term stochastic components, and derive a formula for futures prices. The long-term component is modelled as a Lévy process with increments belonging to the class of generalised hyperbolic distributions. We describe the short-term factor by Lévy semistationary processes: we start from a CARMA(2,1), i.e. a continous-time ARMA model, and generalise it by adding a short-memory stochastic volatility. We further modify the model by including the information about the wind energy production as an exogenous variable. We fit our models to German and Austrian data including spot and futures prices as well as the wind energy production and total load data. Empirical studies reveal that taking into account the impact of the wind energy generation on the prices improves the goodness of fit.

12:15 - 13: 00 Pierre Gruet (EDF R&D), Determining the optimal number of factors in a N-factor price model for electricity


We investigate a class of stochastic differential equations driven by a given number of Brownian motions to model the price of electricity forward contracts. After describing the market, we motivate the use of this class of models and estimate their parameters. This allows us to choose a criterion to compare the models, and we illustrate our results on market data.

13:00 – 14:15 Lunch

14:15 – 15:00 Markus Bibinger (Philipps-Universität Marburg), Volatility estimation for stochastic PDEs using high-frequency observations (joint work with Mathias Trabs)


We study the parameter estimation for parabolic, linear, second order, stochastic partial differential equations (SPDEs) observing a mild solution on a discrete grid in time and space. The SPDE model covers many interesting applications, we highlight its use for term structure models. A high-frequency regime is considered where the mesh of the grid in the time variable goes to zero. Focusing on volatility estimation, we provide an explicit and easy to implement method of moments estimator based on the squared increments of the process. The estimator is consistent and admits a central limit theorem. This is established moreover for the estimation of the integrated volatility in a semi-parametric framework. Starting from a representation of the solution as an infinite factor model and exploiting mixing properties of Gaussian time series, the theory considerably differs from the statistics for semi-martingales literature. The performance of the method is illustrated in a simulation study.

15:00 - 15:45 John Moriarty (Queen Mary University of London), Rare events in energy networks and markets: an MCMC approach


Energy networks and markets experience various types of disturbance. For example, the increasing penetration of renewable energy sources is increasing the variability of power generation, with both physical and financial consequences. Further, unusually large power disturbances propagate in a complex manner due to network effects. We present a novel extension, named ghost sampling, of the Metropolis-Hastings Markov Chain Monte Carlo method that is tailored to efficiently sample rare power disturbances, conditional on some unusual physical or financial outcome. Generating a representative random sample provides insight into the effect of stochastic generation on, among others, locational marginal prices and system security, and we present examples from small simulated networks. Our method can perform conditional sampling from any joint distribution of power disturbances and thus capture important statistical features of renewable generation including, for instance, correlated and non-Gaussian disturbances.

15:45 – 16:15 Coffee Break

16:15 – 17:00 Rafal Weron (Wrocław University of Technology), Recent advances in electricity price forecasting: A 2018 perspective


A variety of methods and ideas have been tried for electricity price forecasting over the last two decades, with varying degrees of success. In this talk I will provide a short overview of the recent advances

17:00  End of the workshop


Registration form