Correlated model fusion. (4th August 2017)
- Record Type:
- Journal Article
- Title:
- Correlated model fusion. (4th August 2017)
- Main Title:
- Correlated model fusion
- Authors:
- Hoegh, Andrew
Leman, Scotland - Other Names:
- Jeske Daniel R. guestEditor.
Xie Min‐ge guestEditor. - Abstract:
- Abstract : Model fusion methods, or more generally ensemble methods, are a useful tool for prediction. Combining predictions from a set of models smooths out biases and reduces variances of predictions from individual models, and hence, the combined predictions typically outperform those from individual models. In many algorithms, individual predictions are arithmetically averaged with equal weights. However, in the presence of correlated models, the fusion process is required to account for association between models; otherwise, the naively averaged predictions will be suboptimal. This article describes optimal model fusion principles and illustrates the potential pitfalls of naive fusion in the presence of correlated models for binary data. An efficient algorithm for correlated model fusion is detailed and applied to algorithms mining social media information to predict civil unrest. Copyright © 2017 John Wiley & Sons, Ltd.
- Is Part Of:
- Applied stochastic models in business and industry. Volume 34:Number 1(2018)
- Journal:
- Applied stochastic models in business and industry
- Issue:
- Volume 34:Number 1(2018)
- Issue Display:
- Volume 34, Issue 1 (2018)
- Year:
- 2018
- Volume:
- 34
- Issue:
- 1
- Issue Sort Value:
- 2018-0034-0001-0000
- Page Start:
- 31
- Page End:
- 43
- Publication Date:
- 2017-08-04
- Subjects:
- ensemble methods -- Gaussian graphical models -- model averaging -- model selection
Stochastic analysis -- Periodicals
Stochastic processes -- Periodicals
Business mathematics -- Periodicals
Finance -- Mathematical models -- Periodicals
Industrial management -- Mathematical models -- Periodicals
338.00151923 - Journal URLs:
- http://onlinelibrary.wiley.com/ ↗
- DOI:
- 10.1002/asmb.2261 ↗
- Languages:
- English
- ISSNs:
- 1524-1904
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 1580.062200
British Library DSC - BLDSS-3PM
British Library STI - ELD Digital store - Ingest File:
- 5828.xml