Agenda and Speakers
Registration & Welcome Coffee
Short break
Strategy Optimization by Means of Evolutionary Algorithms With Multiple Closing Criteria for Energy Trading
Silvia Trimarchi
A2A SpA
Abstract: The energy markets are experiencing an enhanced volatility and unpredictability due to the growing integration of renewable energy sources in the grid and to the unstable geopolitical situation that is developing worldwide. Energy traders are therefore raising concerns on how to achieve solutions that not only ensure stability in terms of energy needs, both on the supply and demand side, but also enable profits within these markets. To cope with the complexity of this emerging scenario, tools that support traders in their decisions, such as algorithmic trading strategies, are attracting always more and more attention. In particular, evolutionary algorithms have emerged as an effective tool for developing robust and innovative trading strategies. Indeed, their flexibility and adaptability allow for the inclusion of various performance metrics. This article employs a recently issued evolutionary algorithm, called social network optimization, to identify the optimal closing criteria of already opened positions in an energy commodity market. More specifically, the proposed trading strategy is based on five self-defined parameters, which determine a profitable solution over nearly six years of available data. In particular, the overall average positive return achieved and the maximum monthly yield of 1.9% highlight the adaptability and robustness of the developed algorithmic trading strategy. Therefore, the results suggest the potentialities of developing and upgrading novel trading strategies by exploiting evolutionary computation techniques in the actual complex energy markets.
Forecast Combination for Tail Risk: the Virtue and Vice of Simple Averaging
Roxana Halbleib
University of Freiburg
Abstract: This paper examines the forecast combination for Value-at-Risk (VaR) and Expected Shortfall (ES). We show that the weighted arithmetic average commonly used to construct a forecast combination utilises the convexity property of the loss function only in case of VaR, while to combine ES one should use the harmonic mean.
To construct combination weights consistent with this aggregation result, we propose a novel forecast combination for tail risk measures based on the Bagged Pretested Forecast Combination (BPFC) algorithm.
The combination weights assigned to candidate models are determined by their predictive performance using the Model Confidence Set (MCS) test. Unlike many traditional combination methods, BPFC adapts to changing market conditions while simultaneously facilitating model selection and improving forecast stability. We evaluate the performance of forecasting combinations for VaR and ES within the framework of consistent loss functions, highlighting the role of convexity in performance improvements. Our results show that the advantages of combining forecasts are especially evident when there is substantial disagreement among candidate models, a situation that commonly arises during turbulent financial periods.
To empirically validate our approach, we apply it to a dataset of 90 stocks spanning various market capitalizations and covering periods of severe financial stress, including the Global Financial Crisis and the COVID-19 pandemic. The results illustrate the ability of BPFC to dynamically select and combine the most effective models from a pool of over 60 candidates, continuously adjusting weights based on model’s forecasting performance and evolving market conditions.
Lunch
From social media to financial market: an exploratory study on sentiment
Gaia Salford
Banco Posta Fondi Spa SGR
Abstract: TBA