Mixing past predictions

Alexander Korotin, Vladimir V’yugin, Evgeny Burnaev
Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications, PMLR 128:171-188, 2020.

Abstract

In the framework of the theory of prediction with expert advice, we present an algorithm for online aggregation of the functional predictions. The approach implies that at each time step some algorithm issues a forecast in the form of a function and then the master algorithm combines these current and past functional forecasts into one aggregated functional forecast. We apply the proposed algorithm for the problem of long-term predictions of time series. By combining the past and current long-term functional forecasts, we obtain a smoothing mechanism that protects our algorithm from temporary changes in the trend of time series, noise and outliers. To evaluate the performance of presented aggregating algorithm as a long-term forecaster we use a new “integral” loss function and the delayed feedback approach. We apply this algorithm for the regression problems, we present some method for smoothing regression forecasts.

Cite this Paper

BibTeX
@InProceedings{pmlr-v128-korotin20a, title = {Mixing past predictions}, author = {Korotin, Alexander and V'yugin, Vladimir and Burnaev, Evgeny}, booktitle = {Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications}, pages = {171--188}, year = {2020}, editor = {Gammerman, Alexander and Vovk, Vladimir and Luo, Zhiyuan and Smirnov, Evgueni and Cherubin, Giovanni}, volume = {128}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v128/korotin20a/korotin20a.pdf}, url = {https://proceedings.mlr.press/v128/korotin20a.html}, abstract = {In the framework of the theory of prediction with expert advice, we present an algorithm for online aggregation of the functional predictions. The approach implies that at each time step some algorithm issues a forecast in the form of a function and then the master algorithm combines these current and past functional forecasts into one aggregated functional forecast. We apply the proposed algorithm for the problem of long-term predictions of time series. By combining the past and current long-term functional forecasts, we obtain a smoothing mechanism that protects our algorithm from temporary changes in the trend of time series, noise and outliers. To evaluate the performance of presented aggregating algorithm as a long-term forecaster we use a new “integral” loss function and the delayed feedback approach. We apply this algorithm for the regression problems, we present some method for smoothing regression forecasts.} }
Endnote
%0 Conference Paper %T Mixing past predictions %A Alexander Korotin %A Vladimir V’yugin %A Evgeny Burnaev %B Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2020 %E Alexander Gammerman %E Vladimir Vovk %E Zhiyuan Luo %E Evgueni Smirnov %E Giovanni Cherubin %F pmlr-v128-korotin20a %I PMLR %P 171--188 %U https://proceedings.mlr.press/v128/korotin20a.html %V 128 %X In the framework of the theory of prediction with expert advice, we present an algorithm for online aggregation of the functional predictions. The approach implies that at each time step some algorithm issues a forecast in the form of a function and then the master algorithm combines these current and past functional forecasts into one aggregated functional forecast. We apply the proposed algorithm for the problem of long-term predictions of time series. By combining the past and current long-term functional forecasts, we obtain a smoothing mechanism that protects our algorithm from temporary changes in the trend of time series, noise and outliers. To evaluate the performance of presented aggregating algorithm as a long-term forecaster we use a new “integral” loss function and the delayed feedback approach. We apply this algorithm for the regression problems, we present some method for smoothing regression forecasts.
APA
Korotin, A., V’yugin, V. & Burnaev, E.. (2020). Mixing past predictions. Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications, in Proceedings of Machine Learning Research 128:171-188 Available from https://proceedings.mlr.press/v128/korotin20a.html.

Related Material