A parallel adaptive metropolis algorithm for uncertainty assessment of Xinanjiang model parameters

Chun Tian Cheng; Xiang Yang Li; Kwok Wing Chau

A parallel adaptive metropolis algorithm for uncertainty assessment of Xinanjiang model parameters

Chun Tian Cheng, Xiang Yang Li, Kwok Wing Chau

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

1 Citation (Scopus)

Abstract

Markov Chain Monte Carlo (MCMC) methods, which are popular for estimating parameter uncertainty of hydrological models, generally converge slowly, and are easy to get stuck in a local optimized region in the parametric space. In this paper we present a new MCMC sampler entitled the Parallel Adaptive Metropolis (PAM) algorithm, which is well suited to estimating the parameter uncertainty of hydrological models. The PAM algorithm provides an adaptive MCMC sampler to estimate the posterior probability distribution of parameters under a Bayesian framework. The performance of the PAM algorithm is greatly improved through parallel computing. The PAM algorithm is applied to assess the parameter uncertainty of the Xinanjiang model using hydrological data from Shuangpai Reservoir, China. The case study demonstrates that there is considerable uncertainty about the Xinanjiang model parameters.

Original language	English
Title of host publication	Hydrological Sciences for Managing Water Resources in the Asian Developing World
Pages	10-16
Number of pages	7
Edition	319
Publication status	Published - 30 Jun 2008
Event	Hydrological Sciences for Managing Water Resources in the Asian Developing World - Guangzhou, China Duration: 8 Jun 2006 → 10 Jun 2006

Conference

Conference	Hydrological Sciences for Managing Water Resources in the Asian Developing World
Country/Territory	China
City	Guangzhou
Period	8/06/06 → 10/06/06

Keywords

Hydrological model
MCMC
PAM
Parallel computing
Uncertainty assessment

ASJC Scopus subject areas

Water Science and Technology
Oceanography

Cite this

@inproceedings{0f00d6d8d25a420b8e1cb4c7ccfe9e7c,

title = "A parallel adaptive metropolis algorithm for uncertainty assessment of Xinanjiang model parameters",

abstract = "Markov Chain Monte Carlo (MCMC) methods, which are popular for estimating parameter uncertainty of hydrological models, generally converge slowly, and are easy to get stuck in a local optimized region in the parametric space. In this paper we present a new MCMC sampler entitled the Parallel Adaptive Metropolis (PAM) algorithm, which is well suited to estimating the parameter uncertainty of hydrological models. The PAM algorithm provides an adaptive MCMC sampler to estimate the posterior probability distribution of parameters under a Bayesian framework. The performance of the PAM algorithm is greatly improved through parallel computing. The PAM algorithm is applied to assess the parameter uncertainty of the Xinanjiang model using hydrological data from Shuangpai Reservoir, China. The case study demonstrates that there is considerable uncertainty about the Xinanjiang model parameters.",

keywords = "Hydrological model, MCMC, PAM, Parallel computing, Uncertainty assessment",

author = "Cheng, {Chun Tian} and Li, {Xiang Yang} and Chau, {Kwok Wing}",

year = "2008",

month = jun,

day = "30",

language = "English",

isbn = "9781901502442",

pages = "10--16",

booktitle = "Hydrological Sciences for Managing Water Resources in the Asian Developing World",

edition = "319",

note = "Hydrological Sciences for Managing Water Resources in the Asian Developing World ; Conference date: 08-06-2006 Through 10-06-2006",

}

A parallel adaptive metropolis algorithm for uncertainty assessment of Xinanjiang model parameters. / Cheng, Chun Tian; Li, Xiang Yang; Chau, Kwok Wing.
Hydrological Sciences for Managing Water Resources in the Asian Developing World. 319. ed. 2008. p. 10-16.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - A parallel adaptive metropolis algorithm for uncertainty assessment of Xinanjiang model parameters

AU - Cheng, Chun Tian

AU - Li, Xiang Yang

AU - Chau, Kwok Wing

PY - 2008/6/30

Y1 - 2008/6/30

N2 - Markov Chain Monte Carlo (MCMC) methods, which are popular for estimating parameter uncertainty of hydrological models, generally converge slowly, and are easy to get stuck in a local optimized region in the parametric space. In this paper we present a new MCMC sampler entitled the Parallel Adaptive Metropolis (PAM) algorithm, which is well suited to estimating the parameter uncertainty of hydrological models. The PAM algorithm provides an adaptive MCMC sampler to estimate the posterior probability distribution of parameters under a Bayesian framework. The performance of the PAM algorithm is greatly improved through parallel computing. The PAM algorithm is applied to assess the parameter uncertainty of the Xinanjiang model using hydrological data from Shuangpai Reservoir, China. The case study demonstrates that there is considerable uncertainty about the Xinanjiang model parameters.

AB - Markov Chain Monte Carlo (MCMC) methods, which are popular for estimating parameter uncertainty of hydrological models, generally converge slowly, and are easy to get stuck in a local optimized region in the parametric space. In this paper we present a new MCMC sampler entitled the Parallel Adaptive Metropolis (PAM) algorithm, which is well suited to estimating the parameter uncertainty of hydrological models. The PAM algorithm provides an adaptive MCMC sampler to estimate the posterior probability distribution of parameters under a Bayesian framework. The performance of the PAM algorithm is greatly improved through parallel computing. The PAM algorithm is applied to assess the parameter uncertainty of the Xinanjiang model using hydrological data from Shuangpai Reservoir, China. The case study demonstrates that there is considerable uncertainty about the Xinanjiang model parameters.

KW - Hydrological model

KW - MCMC

KW - PAM

KW - Parallel computing

KW - Uncertainty assessment

UR - http://www.scopus.com/inward/record.url?scp=45749155302&partnerID=8YFLogxK

M3 - Conference article published in proceeding or book

SN - 9781901502442

SP - 10

EP - 16

BT - Hydrological Sciences for Managing Water Resources in the Asian Developing World

T2 - Hydrological Sciences for Managing Water Resources in the Asian Developing World

Y2 - 8 June 2006 through 10 June 2006

ER -

A parallel adaptive metropolis algorithm for uncertainty assessment of Xinanjiang model parameters

Abstract

Conference

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this