Journal article
Bayesian algorithms for missing observations in experimental designs for a nanolubrication process
IIE transactions, Vol.41(11), pp.969-978
09/01/2009
DOI: 10.1080/07408170902806888
Abstract
Three new Bayesian algorithms for missing observations based on predictive ability and minimization of the Residual Sum of Squares (RSS) are proposed. Their performance is compared to three existing algorithms based on an appropriate predicted residual error sum of squares statistic. Different positions of the missing observations and initial model conditions are considered. In all the investigated cases, the Bayesian algorithms perform significantly better than non-Bayesian algorithms. A numerical study is performed using a nanolubrication process. It shows that the Bayesian complete RSS minimization algorithm yields the closest estimates of the missing observations, with the maximum predictive ability.
Details
- Title: Subtitle
- Bayesian algorithms for missing observations in experimental designs for a nanolubrication process
- Creators
- Navinchandra N Acharya - Harold and Inge Marcus Department of Industrial and Manufacturing Engineering , The Pennsylvania State UniversityHarriet Black Nembhard - Harold and Inge Marcus Department of Industrial and Manufacturing Engineering , The Pennsylvania State University
- Resource Type
- Journal article
- Publication Details
- IIE transactions, Vol.41(11), pp.969-978
- Publisher
- Taylor & Francis Group
- DOI
- 10.1080/07408170902806888
- ISSN
- 0740-817X
- eISSN
- 1545-8830
- Language
- English
- Date published
- 09/01/2009
- Academic Unit
- Engineering Administration; Industrial and Systems Engineering
- Record Identifier
- 9984121864202771
Metrics
9 Record Views