Proximity is measured over citations between two papers we both hold, weighted by how heavily one leans on the other, and is symmetric — it does not distinguish citing from being cited. Authors without a profile here are skipped, and a genuinely close colleague can be missing simply because their work is not in our arXiv corpus. Method: docs/06-citations-pipeline.md.
Postprocessing of point predictions for probabilistic forecasting of day-ahead electricity prices: The benefits of using isotonic distributional regression
Assembled from arXiv and OpenAlex. Duplicate records for the same paper are merged, and the published version is shown where we could identify one. Corrections welcome.