Conformal prediction interval estimation and applications to day-ahead and intraday power markets

In this study, we investigated the application of conformal prediction (CP) concept in context short-term electricity price forecasting. particular, determined most important aspects related to utility CP, as well explaining why simple but highly effective idea has proved useful other areas and its characteristics make it promising for power applications. We compared performance CP with various state-of-the-art forecasting models, such quantile regression averaging, an empirical out-of-sample study three time series. combined underlying point forecast models demonstrate versatility behavior under changing conditions. Our findings suggest that yields sharp reliable intervals markets. also inspected effects each model components provide path-based guideline regarding how find best market.