[Openspace] Multi-Level Modeling
rbanerjee at prev.org
Fri Apr 28 19:06:12 CDT 2006
Guangqing and Jason,
There is software that does hierarchical regression. For example,
WinBUGS(TM) uses Bayesian statistics (involving networks and
hierarchies) to model a variety of problems including space and time
networks. One can also write programs in C++ etc to implement these
Bayes methods. The problem is that convergence is based on algorithms
using Markov Chains methods and are very time consuming. Good Luck.
Prevention Research Center,
>>> GUANGQING CHI <guangqingchi at wisc.edu> 4/28/2006 4:00 PM >>>
As far as I know, no existing software can do spatial hierarchical
regression. You have to conduct the analysis by two stages as
demonstrated in the following paper.
Sampson, Robert J., Jeffrey D. Morenoff & Felton Earls. 1999. "Beyond
Social Capital: Spatial Dynamics of Collective Efficacy for Children."
American Sociological Review 64(5):633-660.
----- Original Message -----
From: Jason McDaniel <jmcdanie at usc.edu>
Date: Friday, April 28, 2006 5:00 pm
Subject: [Openspace] Multi-Level Modeling
To: openspace at sal.uiuc.edu
> I am wondering about the best ways to combine spatial regression
> opinion survey data. I have been thinking about the possibilities
> using a multi-level hierarchical model that combines spatially
> located data of Los Angeles Mayoral elections and demographics
> aggregated at the census tract level with exit polling. Does
> know of any efforts to develop multi-level models using spatial
> Thank you.
> Jason McDaniel
> Department of Political Science
> University of Southern California
> jmcdanie at usc.edu
> Openspace mailing list
> Openspace at sal.uiuc.edu
Openspace mailing list
Openspace at sal.uiuc.edu
More information about the Openspace