Welcome to DREAM: global adaptive MCMC project!
DiffeRential Evolution Adaptive Metropolis (DREAM). Efficient global MCMC even in high-dimensional spaces. Developed by J.A. Vrugt, C.J.F. ter Braak et al.
The project summary page you can find here.
For information on how to use dream, please run in R:
- help("dreamCalibrate") - to calibrate a function using dream
- help("dream") - for low-level interface
- demo(example1) - Fitting a banana shaped distribution
- demo(example2) - Fitting an n-dimensional Gaussian distribution
- demo(FME.nonlinear.model) - Calibrating the non-linear model shown
in the FME package vignette
- demo(FME.nonlinear.model_parallelisation) - Example of parallelisation using the SNOW package
- demo(parallelisation_chain_id) - Example of parallelisation when DREAM calls an external model using batch files in separate folders.
To cite the DREAM algorithm please use:
-
Vrugt, J. A., ter Braak, C. J. F., Diks, C. G. H., Robinson, B. A.,
Hyman, J. M., Higdon, D., 2009. Accelerating Markov chain Monte Carlo
simulation by differential evolution with self-adaptive randomized
subspace sampling. International Journal of Nonlinear Sciences
and Numerical Simulation 10 (3), 273-290. DOI: 10.1515/IJNSNS.2009.10.3.273
To cite the dream package, please use:
For additional information on the algorithm also see:
-
Vrugt, J. A., ter Braak, C. J. F., Gupta, H. V., Robinson, B. A.,
2009. Equifinality of formal (DREAM) and informal (GLUE) Bayesian
approaches in hydrologic modeling?
Stochastic Environmental Research and Risk Assessment 23 (7), 1011--1026.
DOI: 10.1007/s00477-008-0274-y
-
An example of parallelisation of DREAM for the SWAT model (ZIP file) can be downloaded. Thanks to John Joseph for this contribution. This example is documented in the publication:
Joseph, J.F., J.H.A. Guillaume (2013) Using a parallelized MCMC algorithm in R to identify appropriate likelihood functions for SWAT, Environmental Modelling & Software, 46, pp 292-298, DOI: 10.1016/j.envsoft.2013.03.012.
This implementation of DREAM has been tested against the original Matlab implementation. See example1.R and example2.R
Please note that the dream_zs and dream_d algorithms may be superior in your circumstances. These are not implemented in this package. Please read the following references for details:
-
Vrugt, J. A. and Ter Braak, C. J. F. (2011) DREAM(D): an adaptive Markov Chain Monte Carlo simulation algorithm to solve discrete, noncontinuous, and combinatorial posterior parameter estimation problems, Hydrol. Earth Syst. Sci., 15, 3701-3713, DOI: 10.5194/hess-15-3701-2011
-
ter Braak, C. and J. Vrugt (2008). Differential Evolution Markov Chain
with snooker updater and fewer chains. Statistics and Computing 18(4): 435-446 DOI: 10.1007/s11222-008-9104-9
-
Laloy,E., and J.A. Vrugt. 2012. High-dimensional posterior exploration
of hydrologic models using multiple-try DREAM(ZS) and high-performance
computing. Water Resources Research, 48, W0156. DOI 10.1029/2011WR010608