Geometry of Nonlinear Forecast Reconciliation
Nespoli, L., Biswas, A., & Medici, V. (2025). "Geometry of Nonlinear Forecast Reconciliation." arXiv preprint arXiv:2507.22500.
CV
Education
- Pursuing Ph.D in Informatics, Università della Svizzera Italiana (USI), Lugano, Switzerland, 2024-present
- M.Sc. in Statistics, University of Hyderabad, Hyderabad, India, 2021-2023
- B.Sc. in Statistics, Presidency University, Kolkata, India, 2018-2021
Work experience
- 2024-current : PhD Student
- ISAAC, Scuola universitaria professionale della Svizzera italiana (SUPSI)
- Mendrisio, Switzerland
- Supervisor: Dr. Lorenzo Nespoli
- 2024-current: Teaching Assistant
- DTI, Scuola universitaria professionale della Svizzera italiana (SUPSI)
- Viganello, Switzerland
- Course instructors: Prof. Giorgio Corani, Dr. Lorenzo Nespoli, Dr. Marco Forgione
- 2023-2024 : Research Assistant
- ForeML Lab, Sorbonne University
- Online
- Supervisor: Prof. Tanujit Chakraborty
Programming Skills
- Python : advanced
- R : advanced
Packages
- BayesReconPy : A package for Bayesian Forecast Reconciliation in Python
Publications
BayesReconPy: A Python package for forecast reconciliation
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
E‑STGCN: Extreme Spatiotemporal Graph Convolutional Networks for Air Quality Forecasting
Panja, M., Chakraborty, T., Biswas, A., & Deb, S. (2024). "E‑STGCN: Extreme Spatiotemporal Graph Convolutional Networks for Air Quality Forecasting." arXiv preprint arXiv:2411.12258.
Talks
PhD Abroad
Talk at Webinar, Online
Probabilistic Nonlinear Forecast Reconciliation Methods: Projection, Importance Sampling, and Kalman Filtering
Talk at International Symposium on Forecasting (ISF) 2025, Friendship Palace, Haidian dist, Beijing, China
Probabilistic forecast reconciliation with bayesRecon
Talk at IIF Workshop on Open Source Forecasting 2025, Beihang University, Haidian dist, Beijing, China
MSGNet: Learning Multi-Scale Inter-Series Correlations for Multivariate Time Series Forecasting
Course presentation at Viganello Campus, Universita Della Svizzera Italiana, Switzerland
Bidirectional LSTM for time-series forecasting: A case study on Ozone data of Ticino, Switzerland
Course presentation at Viganello Campus, Universita Della Svizzera Italiana, Switzerland
xLSTM: Extended Long Short-Term Memory
Course presentation at Viganello Campus, Universita Della Svizzera Italiana, Switzerland
Lebesgue’s Criterion for Riemann-integrability
Course presentation at Asynchronous, University of Hyderabad, Hyderabad, India
Linear Rank Statistic and General two-sample problem
Course presentation at Statistics Department, Presidency University, Kolkata, India
A brief introduction to Indian Official Statistics and discussion on the role of MoSPI, CSO and NSSO
Course presentation at Statistics Department, Presidency University, Kolkata, India
Teaching
🎓 Teaching content coming soon.