\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}
\usepackage{url}
\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, August 16, 2005\\
  Sequoia Hall Room 200\\
  (Cookies at 3:45 in 1st Floor Lounge)
\end{center}

\begin{center}
  \textsl{Carlos Rivero} \\
  Universidad Complutense de Madrid
\end{center}
\begin{center}
  \textbf{Linearization procedures to test for breaks in nonlinear models}
\end{center}


Abstract:\\

Testing for structural breaks in nonlinear and dynamic models might be
quite involving. This seminar presents some easy to implement two-step
procedures to test for constant parameters in nonlinear and dynamic
models using a Wald type test and the Andrews sup Wald test. The first
step obtains full sample consistent estimates under the null, and then
the model is linearized. In the second step, the linearized model is
recursively estimated and tested. Introducing the uniform consistency
property, it is proved that the asymptotic distributions of the
statistics to test for the existence of breaks in the linearized model
are the same as those obtained for the nonlinear model. A Monte Carlo
simulation example supports the relevance of the theory and shows the
performance of the proposed procedures.

Joint work with Juan del Hoyo and Guillermo Llorente.
\end{document}