The design of electricity imbalance pricing mechanisms is internationally controversial. Policies on whether to permit virtual trading for market participants and whether, or how, to impose penalty incentives on the imbalance volumes vary widely. Furthermore, market designs vary depending whether the imbalance prices are obtained directly from real-time trading or based upon the offer and demand functions from the day-ahead energy markets. This paper develops an analytical framework for evaluating designs for imbalance settlement mechanisms and we have selected the Japanese electricity market, which has undergone several revisions in its imbalance mechanism, as a good example to assess such variations. We develop a predictive approach for the imbalance volumes and price densities using two-step quantile regressions and derive a new trading optimization for a virtual trader's arbitrage position. We construct supporting models to estimate prediction errors for renewable power and demand as drivers of imbalance volume. The empirical analysis reveals that even in a mechanism with imbalance penalties based upon day-ahead reference prices, virtual trading may still be beneficial to market participants as well as to the system operator. We also find that greater market transparency is crucial for increased benefits. The insights generalize beyond the Japan case study.

Short-term risk management is becoming increasingly significant in power trading as the intermittent renewable generators introduce more weather risk into the price formation dynamics. There is a vacuum in hedging instruments at the day-ahead stage to protect retailers in particular from such volatility and price spikes. Motivated by this requirement, this paper analyses a flexible hedging product, day-ahead cap futures. For pricing this product, we parametrically predict the probability distribution of day-ahead prices using the multifactor Generalized Additive Model for Location, Scale and Shape (GAMLSS) based upon the skew-t distribution with weather forecasts and calendar information as explanatory variables. In particular, we reveal that this higher-order moment model is superior to several lower-order models such as the normal distribution in all the following three aspects: fairness as pricing method, underwriting risk of the risk-taker and the variance reduction effect of the risk hedger.

Due to the inherent uncertainty of wind conditions as well as the price unpredictability in the competitive electricity market, wind power producers are exposed to the risk of concurrent fluctuations in both price and volume. Therefore, it is imperative to develop strategies to effectively stabilize their revenues, or cash flows, when trading wind power output in the electricity market. In light of this context, we present a novel endeavor to construct multivariate derivatives for mitigating the risk of fluctuating cash flows that are associated with trading wind power generation in electricity markets. Our approach involves leveraging nonparametric techniques to identify optimal payoff structures or compute the positions of derivatives with fine granularity, utilizing multiple underlying indexes including spot electricity price, area-wide wind power production index, and local wind conditions. These derivatives, referred to as mixed derivatives, offer advantages in terms of hedge effectiveness and contracting efficiency. Notably, we develop a methodology to enhance the hedge effects by modeling multivariate functions of wind speed and wind direction, incorporating periodicity constraints on wind direction via tensor product spline functions. By conducting an empirical analysis using data from Japan, we elucidate the extent to which the hedge effectiveness is improved by constructing mixed derivatives from various perspectives. Furthermore, we compare the hedge performance between high-granular (hourly) and low-granular (daily) formulations, revealing the advantages of utilizing a high-granular hedging approach.

In recent years, as photovoltaic (PV) power generation has rapidly increased on a global scale, there is a growing need for a highly accurate power generation forecasting model that is easy to implement for a wide range of electric utilities. Against this background, this study proposes a PV power forecasting model based on the generalized additive model (GAM) and compares its forecasting accuracy with four popular machine learning methods: k-nearest neighbor, artificial neural networks, support vector regression, and random forest. The empirical analysis provides an intuitive interpretation of the multidimensional smooth trends estimated by the GAM as tensor product splines and confirms the validity of the proposed modeling structure. The effectiveness of GAM is particularly evident in trend completion for missing data, where it is able to flexibly express the tangled trend structure inherent in time series data, and thus has an advantage not only in interpretability but also in improving forecast accuracy.