Aggregation of Adaptive Forecasting Algorithms Under Asymmetric Loss Function
The paper deals with applying the strong aggregating algorithm to games with asymmetric loss function. A particular example of such games is the problem of time series forecasting where specific losses from under-forecasting and over-forecasting may vary considerably. We use the aggregating algorithm for building compositions of adaptive forecasting algorithms. The paper specifies sufficient conditions under which a composition based on the aggregating algorithm performs as well as the best of experts. As a result, we find a theoretical bound for the loss process of a given composition under asymmetric loss function. Finally we compare the composition based on the aggregating algorithm to other well-known compositions in experiments with real data.
KeywordsAggregating algorithm Time series forecasting Asymmetric loss function
Unable to display preview. Download preview PDF.
- 1.Vovk, V.: Competitive on-line statistics. International Statistical Review 69 (2001)Google Scholar
- 4.Vovk, V., Watkins, C.: Universal portfolio selection. In: Proceedings of the Eleventh Annual Conference on Computational Learning Theory, COLT, pp. 12–23 (1998)Google Scholar
- 6.Lukashin, Y.P.: Adaptive methods of short-term time series forecasting. Financy i statistica 415, Moscow (2003) (in Russian)Google Scholar
- 7.Romanenko, A.A.: Aggregation of adaptive forecast algorithms. Reports of the 15th Russian Conference on Mathematical Methods of Image Recognition, pp. 170–173, Moscow (2011) (in Russian)Google Scholar
- 11.Timmermann, A.G.: Forecast Combinations. CEPR Discussion Papers (2006). http://ideas.repec.org/p/cpr/ceprdp/5361.html
- 12.Postnikova, E.: Quantile Regression. NSU 34 (2000) (in Russian)Google Scholar