Hi:
I am running some standard CAR models in Winbugs1.4. For practice, I am
using 6 years of vandalism data for about 100 neighborhoods. I have
some covariates too, but they are available only for one or two recent
years. I wanted to link some neighborhood characteristics to the count
of vadalism. I am using the negative binomial distribution.
I aggregated all the years and ran the CAR model assuming that the
characteristics of the neighborhood have not changed.
The basic descriptive stat shows that the number of neighborhoods
affected by vadalism incidents increase over time in adramatic fashion
covering a larger area by the end of the latest year.
What would be the best way to model this to show that the crime
coverage (spillover effect) is has changed over time --- in a CA
framework? What statistics should I be looking at or calculate? The
only thing I have is the spatial heterogeneity S[i] series I get and
the corresponding std.
I though about running year by year CAR model and show some sort of
Moran-type statistics? Is that the way to go? If so, how can I
calculate the Moran stat in Winbugs?
Any help or code will be highly appreciated (I have created the weight
matrix by counting the number of adjoining neighborhoods etc.)
Is it possible to calculate the probability of vandalism for each
neighborhood? I want to show how these probabilities change over time
for some key neoghborhoods.
Thanks
Alok K. Bohara, Ph.D.
Professor
Department of Economics
University of New Mexico
Albuquerque, NM 87131,USA
ph: 505-277-5903/5304(w)
fax:505-277-9445
[log in to unmask]
http://www.unm.edu/~econ/faculty/professors.html
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|