Title :
Experimental investigation of the efficiency of universal measures based forecasting methods
Author :
Pristavka, Pavel ; Ryabko, Boris
Author_Institution :
Siberian State Univ. of Telecommun. & Inf. Sci., Novosibirsk, Russia
Abstract :
We describe and experimentally investigate a method to construct forecasting algorithms for stationary and ergodic processes based on universal measures (or so-called universal data compressors). Using the time series of solar activity indices as a research target, we show that the quality of the predictions is higher than that of known methods forecasts.
Keywords :
forecasting theory; statistical mechanics; time series; ergodic processes; forecasting algorithms; solar activity indices; stationary processes; time series; universal data compressors; universal measures; Control systems; Encoding; Estimation; Forecasting; Redundancy; Time series analysis; Weather forecasting; Time series; nonparametric methods; solar flux; universal coding; universal measure;
Conference_Titel :
Problems of Redundancy in Information and Control Systems (RED), 2012 XIII International Symposium on
Conference_Location :
St. Petersburg
Print_ISBN :
978-1-4673-0168-8
DOI :
10.1109/RED.2012.6338409
