Bagging and the Bayesian Bootstrap
Publication
, Journal Article
Clyde, M; Lee, HK
Published in: Artificial Intelligence and Statistics
2001
Bagging is a method of obtaining more ro- bust predictions when the model class under consideration is unstable with respect to the data, i.e., small changes in the data can cause the predicted values to change significantly. In this paper, we introduce a Bayesian ver- sion of bagging based on the Bayesian boot- strap. The Bayesian bootstrap resolves a the- oretical problem with ordinary bagging and often results in more efficient estimators. We show how model averaging can be combined within the Bayesian bootstrap and illustrate the procedure with several examples.
Duke Scholars
Published In
Artificial Intelligence and Statistics
Publication Date
2001
Volume
8
Start / End Page
169 / 174
Publisher
Morgan Kaufman Publishers
Citation
APA
Chicago
ICMJE
MLA
NLM
Clyde, M., & Lee, H. K. (2001). Bagging and the Bayesian Bootstrap. Artificial Intelligence and Statistics, 8, 169–174.
Clyde, M., and H. K. Lee. “Bagging and the Bayesian Bootstrap.” Edited by T. Richardson and T. Jaakkola. Artificial Intelligence and Statistics 8 (2001): 169–74.
Clyde M, Lee HK. Bagging and the Bayesian Bootstrap. Richardson T, Jaakkola T, editors. Artificial Intelligence and Statistics. 2001;8:169–74.
Clyde, M., and H. K. Lee. “Bagging and the Bayesian Bootstrap.” Artificial Intelligence and Statistics, edited by T. Richardson and T. Jaakkola, vol. 8, Morgan Kaufman Publishers, 2001, pp. 169–74.
Clyde M, Lee HK. Bagging and the Bayesian Bootstrap. Richardson T, Jaakkola T, editors. Artificial Intelligence and Statistics. Morgan Kaufman Publishers; 2001;8:169–174.
Published In
Artificial Intelligence and Statistics
Publication Date
2001
Volume
8
Start / End Page
169 / 174
Publisher
Morgan Kaufman Publishers