An Efficient Two-Stage Markov Chain Monte Carlo Method for Dynamic Data Integration. Efendiev, Y., Datta-Gupta, A., Ginting, V., Ma, X., & Mallick, B. Water Resources Research, 41(12):n/a–n/a, 2005. W12423
An Efficient Two-Stage Markov Chain Monte Carlo Method for Dynamic Data Integration [link]Paper  abstract   bibtex   
In this paper, we use a two-stage Markov chain Monte Carlo (MCMC) method for subsurface characterization that employs coarse-scale models. The purpose of the proposed method is to increase the acceptance rate of MCMC by using inexpensive coarse-scale runs based on single-phase upscaling. Numerical results demonstrate that our approach leads to a severalfold increase in the acceptance rate and provides a practical approach to uncertainty quantification during subsurface characterization.
@article {WRCR:WRCR10215,
author = {Efendiev, Y. and Datta-Gupta, A. and Ginting, V. and Ma, X. and Mallick, B.},
title = {An {E}fficient {T}wo-{S}tage Markov {C}hain Monte Carlo {M}ethod for {D}ynamic {D}ata {I}ntegration},
journal = {Water Resources Research},
volume = {41},
number = {12},
issn = {1944-7973},
url = {http://dx.doi.org/10.1029/2004WR003764},
pages = {n/a--n/a},
keywords = {Uncertainty assessment, Uncertainty quantification, Inverse theory, MCMC, upscaling, acceptance rate},
year = {2005},
note = {W12423},
abstract="In this paper, we use a two-stage Markov chain Monte Carlo (MCMC) method for subsurface characterization that employs coarse-scale models. The purpose of the proposed method is to increase the acceptance rate of MCMC by using inexpensive coarse-scale runs based on single-phase upscaling. Numerical results demonstrate that our approach leads to a severalfold increase in the acceptance rate and provides a practical approach to uncertainty quantification during subsurface characterization."
}

%-------------------------2004-------------------------------
Downloads: 0