Conformal time series decomposition with component-wise exchangeability

Derck Prinzhorn, Thijmen Nijdam, Putri Van der Linden, Alexander Timans
Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications, PMLR 230:432-465, 2024.

Abstract

Conformal prediction offers a practical framework for distribution-free uncertainty quantification, providing finite-sample coverage guarantees under relatively mild assumptions on data exchangeability. However, these assumptions cease to hold for time series due to their temporally correlated nature. In this work, we present a novel use of conformal prediction for time series forecasting that incorporates time series decomposition. This approach allows us to model different temporal components individually. By applying specific conformal algorithms to each component and then merging the obtained prediction intervals, we customize our methods to account for the different exchangeability regimes underlying each component. Our decomposition-based approach is thoroughly discussed and empirically evaluated on synthetic and real-world data. We find that the method provides promising results on well-structured time series, but can be limited by factors such as the decomposition step for more complex data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v230-prinzhorn24a, title = {Conformal time series decomposition with component-wise exchangeability}, author = {Prinzhorn, Derck and Nijdam, Thijmen and Van der Linden, Putri and Timans, Alexander}, booktitle = {Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications}, pages = {432--465}, year = {2024}, editor = {Vantini, Simone and Fontana, Matteo and Solari, Aldo and Boström, Henrik and Carlsson, Lars}, volume = {230}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v230/main/assets/prinzhorn24a/prinzhorn24a.pdf}, url = {https://proceedings.mlr.press/v230/prinzhorn24a.html}, abstract = {Conformal prediction offers a practical framework for distribution-free uncertainty quantification, providing finite-sample coverage guarantees under relatively mild assumptions on data exchangeability. However, these assumptions cease to hold for time series due to their temporally correlated nature. In this work, we present a novel use of conformal prediction for time series forecasting that incorporates time series decomposition. This approach allows us to model different temporal components individually. By applying specific conformal algorithms to each component and then merging the obtained prediction intervals, we customize our methods to account for the different exchangeability regimes underlying each component. Our decomposition-based approach is thoroughly discussed and empirically evaluated on synthetic and real-world data. We find that the method provides promising results on well-structured time series, but can be limited by factors such as the decomposition step for more complex data.} }
Endnote
%0 Conference Paper %T Conformal time series decomposition with component-wise exchangeability %A Derck Prinzhorn %A Thijmen Nijdam %A Putri Van der Linden %A Alexander Timans %B Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications %C Proceedings of Machine Learning Research %D 2024 %E Simone Vantini %E Matteo Fontana %E Aldo Solari %E Henrik Boström %E Lars Carlsson %F pmlr-v230-prinzhorn24a %I PMLR %P 432--465 %U https://proceedings.mlr.press/v230/prinzhorn24a.html %V 230 %X Conformal prediction offers a practical framework for distribution-free uncertainty quantification, providing finite-sample coverage guarantees under relatively mild assumptions on data exchangeability. However, these assumptions cease to hold for time series due to their temporally correlated nature. In this work, we present a novel use of conformal prediction for time series forecasting that incorporates time series decomposition. This approach allows us to model different temporal components individually. By applying specific conformal algorithms to each component and then merging the obtained prediction intervals, we customize our methods to account for the different exchangeability regimes underlying each component. Our decomposition-based approach is thoroughly discussed and empirically evaluated on synthetic and real-world data. We find that the method provides promising results on well-structured time series, but can be limited by factors such as the decomposition step for more complex data.
APA
Prinzhorn, D., Nijdam, T., Van der Linden, P. & Timans, A.. (2024). Conformal time series decomposition with component-wise exchangeability. Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications, in Proceedings of Machine Learning Research 230:432-465 Available from https://proceedings.mlr.press/v230/prinzhorn24a.html.

Related Material