(Page 1 of 26) - A Conditional Probability Analysis of Pattern-Based Models Applied to Event Data in the Israeli-Palestinian Conflict authored by Schrodt, Philip. and Hudson, Valerie.

A Conditional Probability Analysis of Pattern-Based Models Applied to

Event Data in the Israeli-Palestinian Conflict

Philip A. Schrodt, University of Kansas

Valerie M. Hudson, Brigham Young University

Version 1.0

27 August 2006

Paper prepared for delivery at the Annual Meeting of the American Political Science Association,
Philadelphia, 31 August-3 September 2006.

Authors can be contacted via email at the following addresses:

## email not listed ##

and

## email not listed ##

This research was supported in part by grants from the U.S. National Science Foundation (SES-
0455158),and the Brigham Young University College of Family, Home, and Social Sciences.

A link to the analytical web site for this project, which includes the graphic tools for analyzing event
patterns can be found at the NKSS project web site http://www.nkss.org. The data set
discussed in this paper, as well as a pdf version of the paper with color graphics, can be downloaded
from the KEDS project web site: http://www.ku.edu/~keds/.

Abstract

Existing formal models of political behavior have followed the lead of the natural sciences and
generally focused on methods that use continuous-variable mathematics. Stephen Wolfram has
recently produced an extended critique of that approach in the natural sciences, and suggested that a
great deal of natural behavior can be accounted for using rules that involve discrete patterns. Over
the past three years we have developed software to display the presence of patterns in event data.
This paper extends our earlier work, which focused on the presence of specific patterns over time, to
look at the stochastic characteristics of these pattern. We are specifically interested in the
relationship between patterns as reflected in their conditional probabilities: does the probability of a
pattern increase or decrease the probability of other patterns? Using data from the Israeli-
Palestinian conflict for the period April 1979 to October 2005, and using some of the common
patterns that we identified in our earlier research, we find that these conditional relationship do hold,
and are almost always positive. As we expected, and consistent with our earlier studies, the strength
of the relationships varies over time and these variations are usually strongly correlated with the
broader qualitative characteristics of the conflict, for example major conflict phases such as the first
and second intifadas and the Oslo negotiation process.