Loading...
Loading...

Go to the content (press return)

Linear compositional trend for the frequency of ocean wave events. A Bayesian approach

Author
Ortego, M.I.; Corral, J.; Egozcue, J. J.; Graffelman, J.
Type of activity
Presentation of work at congresses
Name of edition
The 19th Annual Conference of the International Association for Mathematical Geosciences
Date of publication
2018
Presentation's date
2018-09-04
Book of congress proceedings
IAMG 2018: 19th Annual Conference of the International Association for Mathematical Geosciences: Olomouc, Czech Republic: september 2-8, 2018: book of abstracts
First page
70
Last page
71
URL
http://www.iamg.org Open in new window
Abstract
The yearly frequency of classes of ocean wave storms in an off shore site is assumed to be a multinomial sample. The multinomial probabilities are assumed to be a time evolving composition, possibly due to climatic trends. Compositional techniques allow to represent compositions in isometric log-ratio-coordinates. These coordinates are modelled as multivariate normal distributed. In order to simplify the model, the asymptotic distribution of the coordinates coming from multinomial counts is used...
Keywords
Bayesian, compositional data analysis, extremes, waves
Group of research
COSDA-UPC - COmpositional and Spatial Data Analysis

Participants