Λέσχη Φίλων Στατιστικής - GrStats forum
AUEB STATS SEMINARS 6/7/2018: The elliptically symmetric angular Gaussian distribution for modeling spherical data by M.Tsagris Forumgrstats

Join the forum, it's quick and easy

Λέσχη Φίλων Στατιστικής - GrStats forum
AUEB STATS SEMINARS 6/7/2018: The elliptically symmetric angular Gaussian distribution for modeling spherical data by M.Tsagris 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 : 957
Join date : 2009-10-21
http://stat-athens.aueb.gr/~grstats/

AUEB STATS SEMINARS 6/7/2018: The elliptically symmetric angular Gaussian distribution for modeling spherical data by M.Tsagris Empty AUEB STATS SEMINARS 6/7/2018: The elliptically symmetric angular Gaussian distribution for modeling spherical data by M.Tsagris

Tue 12 Jun 2018 - 17:20
AUEB STATS SEMINARS 6/7/2018: The elliptically symmetric angular Gaussian distribution for modeling spherical data by M.Tsagris Tsagri10

AUEB STATISTICS SEMINAR SERIES JULY 2018

Mihalis Tsagris
Postdoctoral researcher in Bioinformatics, Department of Computer Science, University of Crete, Greece

The elliptically symmetric angular Gaussian distribution for modeling spherical data

FRIDAY 6/7/2018
13:00 – 15:00

ROOM Τ103, 1st FLOOR,
NEW AUEB BUILDING (TROIAS 2)

ABSTRACT

Directional data arise in many fields including biology, ecology gelogy, geophysics, criminology, medicine.  When dealing with spherical data, the angular Gaussian (AG) distribution on the sphere Sd-1 is analogous to the Fisher-Bingham (FB) distribution on Sd-1. The Kent distribution is an important subfamily of the general FB distribution which has elliptical contours. In this talk I will present an analogous subfamily of the general AG distribution, called elliptically symmetric angular Gaussian (ESAG) distribution. The ESAG distribution is particularly easy to simulate and has a density that is quick to evaluate exactly. I also discuss the use of ESAG for regression modelling on the sphere. The numerous advantages it offers include spherical-spherical regression and general spherical regression without assuming rotational symmetry (the analogue of independence in the bivariate normal in R2).

Facebook event: https://www.facebook.com/events/463100870808534/
Back to top
Permissions in this forum:
You cannot reply to topics in this forum