BayesReconPy: A Python package for forecast reconciliation
Published in Journal of Open Source Software, 2025
Authors: Anubhab Biswas, Lorenzo Nespoli, Dario Azzimonti, Lorenzo Zambon, Nicolò Rubattu, Giorgio Corani
BayesReconpy implements methods for probabilistic forecast reconciliation in Python. It reconciles hierarchies containing real-valued time series, discrete time series and a mixture of real-valued and discrete time series (mixed hierarchies).
Read the full paper on JOSS. Source code available on GitHub.
Recommended citation: Biswas, A., Nespoli, L., Azzimonti, D., Zambon, L., Rubattu, N., & Corani, G. (2025). "BayesReconPy: A Python package for forecast reconciliation." Journal of Open Source Software, 10(111), 8336. https://doi.org/10.21105/joss.08336
Download Paper | Download Bibtex