Input Selection Based on Information Theory for Constructing Predictor Models of Solar and Geomagnetic Activity Indices

Details Input Selection Based on Information Theory for Constructing Predictor Models of Solar and Geomagnetic Activity Indices Input Selection Based on Information Theory for Constructing Predictor Models of Solar and Geomagnetic Activity Indices

Sep 2009

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2009soph..258..297r&link_type=abstract

Solar Physics, Volume 258, Issue 2, pp.297-318

Physics

1

Input Variables Selection, Mutual Information, Solar And Geomagnetic Indices, Time Series Prediction

Scientific paper

Various phenomena with solar origin and their mutual dependence must be studied in order to predict behaviors in solar - terrestrial system. Linear statistical methods prevalent in analyzing natural systems may not be able to detect nonlinear dependencies among solar and geomagnetic processes. When relations, whether linear or nonlinear, between indices and their changes over time are revealed, better predictions can be made through appropriate modeling techniques. Selection of nonredundant input variables to build suitable models for prediction of solar and geomagnetic activity is of utmost importance. Mutual information is a tool that is capable of capturing all dependencies for detecting nonlinear relations and selecting the best subset of input variables by means of an applicable algorithm that maximizes information about the output and minimizes the shared information between inputs. High generalization power and improved interpretability of the selected inputs are the consequences of this analysis.

Affiliated with

Also associated with

No associations

Rating

Input Selection Based on Information Theory for Constructing Predictor Models of Solar and Geomagnetic Activity Indices does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Input Selection Based on Information Theory for Constructing Predictor Models of Solar and Geomagnetic Activity Indices, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Input Selection Based on Information Theory for Constructing Predictor Models of Solar and Geomagnetic Activity Indices will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1074930