\magnification=1200
\baselineskip=20pt
\nopagenumbers
\font\big=cmr12 scaled \magstep2

\centerline{\bf STANFORD UNIVERSITY}
\centerline{\bf DEPARTMENT OF STATISTICS}
\centerline{\big DEPARTMENTAL SEMINAR}
\bigskip

\baselineskip=12pt
\centerline{4:15 p.m., Tuesday, June 6, 2000}
\centerline{Sequoia Hall Rm. 200}
\centerline{(Cookies at 3:45 in 1st Floor Lounge)}

\bigskip
\baselineskip=15pt
\centerline{\sl Rainer von Sachs}
\centerline{\sl Universite Catholique de Louvain, Belgium}
\bigskip
\centerline{\bf A SLEX-Wavelet model for spectral analysis of bivariate nonstationary time series}
\bigskip

We propose a new method for analyzing bivariate nonstationary time series
which is based on the SLEX (Smooth Localized Complex EXponential)
transform, a library of orthonormal complex-valued transforms.
Simultaneously this method segments the data into approximately stationary
blocks and provides for automatic smoothing of spectral and cross spectral
estimates at each block by GCV. 

By work in progress we further propose a new SLEX-Wavelet model of
nonstationary processes
 including a time-varying spectrum which allows us to develop asymptotic
theory on the consistency of our SLEX estimation method.

Our analysis is motivated from application to an EEG data set recorded
during an ellipetic seizure.

This work is joint with H. Ombao, U Pittsburgh, B. Malow, UMICH Ann Arbor,
Wensheng Guo and the late Jonathan Raz, UPENN, Philadelphia.

\bye