\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, February 25, 2003}
\centerline{Sequoia Hall Room 200}
\centerline{(Cookies at 3:45 in 1st Floor Lounge)}

\bigskip
\baselineskip=15pt
\centerline{\sl Terry O'Neill}
\centerline{\sl Australian National University}
\bigskip
\centerline{\bf Truncated Regression Models}
\bigskip

Motor vehicle accident data is usually truncated. Many of the most
important
databases only have information on accidents involving fatalities. So an
accident will appear on the database if and only if at least one of the
occupants of the vehicle dies in the accident. In the extreme, if a
safety
measure provides perfect protection, then no vehicles with that safety
feature will appear in the database. Techniques such as Conditional
Logisitic Regression can be used to assess the effect of factors such as
seating position within a vehicle, but may be of no use if all the car
occupants have the same level of the factor, for example the speed of
the
vehicle. This talk discusses various methods for the regression analysis
of
truncated regression data, including the estimation of the regression
parameters as well as the distribution of the covariates. Applications
to motor vehicle accident data and wildlife research are discussed.


\bye