Posted by: Janine | February 19, 2011

Statistics

Multivariate
Spatial Poisson Mixtures Are
My Bitch Forever

Dan Gillis

University of Guelph

I present a new statistical method to classify spatially correlated data into distinct groups, while estimating the effect of covariates, using a Mixture model with multivariate conditionally autoregressive random effects. The method provides parameter estimates as good or better than traditional spatial methods, while at the same time classifying the data into distinct groups; an option unavailable to traditional spatial methods. The method was applied to Gastrointestinal data which were classified as either foodborne or waterborne in nature.

Advertisements

Responses

  1. […] morning I awoke to the following email in my inbox: Dan, Nice work! I posted the first one at https://dissertationhaiku.wordpress.com/2011/02/19/statistics-4/ I’ll post the second one in a few days if that’s cool. I like them both […]

  2. […] I present a new statistical method to classify spatially correlated data into distinct groups, while estimating the effect of covariates, using a Mixture model with multivariate conditionally autoregressive random effects.  The method provides parameter estimates as good or better than traditional spatial methods, while at the same time classifying the data into distinct groups; an option unavailable to traditional spatial methods.  The method was applied to Gastrointestinal data which were classified as either foodborne or waterborne in nature.  See also this haiku. […]

  3. […] The first haiku, for those who may have forgotten, can be found here. […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Categories

%d bloggers like this: