\documentclass[11pt]{article}
\setlength{\oddsidemargin}{0.0truein}
\setlength{\evensidemargin}{0.0truein}
\setlength{\textwidth}{6.5truein}
\setlength{\topmargin}{0.0truein}
\setlength{\textheight}{9.0truein} \setlength{\headsep}{0.0truein}
\setlength{\headheight}{0.0truein} \setlength{\topskip}{10.0pt}
\setlength{\parskip}{5mm}
\usepackage{url}
\usepackage{amsmath}
\usepackage{amssymb}
\begin{document}

\begin{center}
  \textbf{\textsc{STANFORD UNIVERSITY}}\\[5pt]
  \textbf{\textsc{DEPARTMENT OF STATISTICS}}\\[5pt]
  \Large{\textbf\textsc{{DEPARTMENTAL SEMINAR}}}
\end{center}
\begin{center}
  4:15 p.m., Tuesday, February 13, 2007\\
  Sequoia Hall Room 200\\
  (Cookies at 3:45 in 1st Floor Lounge)
\end{center}


\begin{center}
  \textsl{Timothy C. Haas} \\
  School of Business Administration\\
University of Wisconsin at Milwaukee
\end{center}
\begin{center}
  \textbf{How to Estimate a Political-Ecological Process Model So That It Can Be Used to Manage an Ecosystem}
\end{center}

\noindent

A predictive understanding of the political processes that produce
a particular sequence of ecosystem management decisions would
provide a tool to gauge how much social change would be needed to
make an ecosystem sustainable. To this end, a stochastic, temporal
model of how political processes influence and are influenced by
ecosystem processes is being developed. This model is realized in
a set of interacting Influence Diagrams (Bayes Nets with Decision
nodes) that each model the decision making of political groups in
countries that affect an ecosystem. These group models also
interact with a model of the affected ecosystem.

As an example, a model has been constructed for the management of
the endangered Cheetah (Acinonyx jubatus) across Kenya, Tanzania,
and Uganda. Here, the ecosystem model is a system of stochastic
differential equations that model cheetah population dynamics. A
data set has been collected that consists of political actions by
groups in these countries along with cheetah counts by political
region.

A frequentist approach to fitting the political-ecological model
to this data is described. For group models, the idea is to
minimize the Hellinger distance between Influence Diagram
multivariate distributions defined by a set of point parameter
values (to represent substantive theory constraints), and the
fitted model { while also maximizing the agreement of the fitted
models' decisions with those observed. For the cheetah count
model, the idea is to also minimize the Hellinger distance to a
theory-based set of parameter values { but to also minimize the
Hellinger distance between the fitted model, and a nonparametric
estimate of the cheetah count's density function.
\end{document}