Λέσχη Φίλων Στατιστικής - GrStats forum
AUEB STATS SEMINARS 18/9/2019: Bayesian analysis of multifidelity computer models with local features and non-nested experimental designs: Application to the WRF model by Georgios Karagiannis Forumgrstats

Join the forum, it's quick and easy

Λέσχη Φίλων Στατιστικής - GrStats forum
AUEB STATS SEMINARS 18/9/2019: Bayesian analysis of multifidelity computer models with local features and non-nested experimental designs: Application to the WRF model by Georgios Karagiannis Forumgrstats
Λέσχη Φίλων Στατιστικής - GrStats forum
Would you like to react to this message? Create an account in a few clicks or log in to continue.
Για προβλήματα εγγραφής και άλλες πληροφορίες επικοινωνήστε με : grstats.forum@gmail.com ή grstats@stat-athens.aueb.gr

Go down
grstats
grstats
Posts : 856
Join date : 2009-10-21
http://stat-athens.aueb.gr/~grstats/

AUEB STATS SEMINARS 18/9/2019: Bayesian analysis of multifidelity computer models with local features and non-nested experimental designs: Application to the WRF model by Georgios Karagiannis Empty AUEB STATS SEMINARS 18/9/2019: Bayesian analysis of multifidelity computer models with local features and non-nested experimental designs: Application to the WRF model by Georgios Karagiannis

Mon 16 Sep 2019 - 10:25
AUEB STATISTICS SEMINAR SERIES
SEPTEMBER 2019

AUEB STATS SEMINARS 18/9/2019: Bayesian analysis of multifidelity computer models with local features and non-nested experimental designs: Application to the WRF model by Georgios Karagiannis 2019_k11



Georgios Karagiannis
Assistant Professor in Statistic, Durham University, Department of Mathematical Sciences

Bayesian analysis of multifidelity computer models with local features and non-nested experimental designs: Application to the WRF model


WEDNESDAY 18/9/2019
13.15-14.30

Room Τ203,  
Troias 2, New AUEB Building

ABSTRACT

Co-kriging is an established framework for the statistical analysis of expensive computer models running at different fidelity levels. Motivated by a Weather Research and Forecasting (WRF) climate model with different resolutions, we develop a new Bayesian treed co-kriging model. The proposed method, unlike existing ones, can take into account local features and discrepancies, while it can be used with non-nested experimental designs. Our procedure utilizes binary treed partition ideas that allow input dependent discrepancies, representation of local features, and discovery of sudden changes in the multifidelity setting. To facilitate the parameter and predictive inference, we design a reversible jump MCMC sampler tailored to the proposed model, which involves collapsed blocks, and direct simulation from conditional distributions. The good performance of our method is demonstrated on artificial benchmark examples and compared against existing methods. The proposed method is implemented for the analysis of a large-scale climate modeling application which involves the WRF model.

The personal web-page and CV of the speaker is available http://www.maths.dur.ac.uk/~mffk55/
Back to top
Permissions in this forum:
You cannot reply to topics in this forum