BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
PRODID:-//Virginia Tech//VTCalendar//EN
BEGIN:VEVENT
DTSTAMP:20121004T193000Z
UID:1348859981969@www.calendar.vt.edu
CATEGORIES:Seminar/Conference
DTSTART:20121004T193000Z
DTEND:20121004T203000Z
SUMMARY:Statistics Thursday Colloquium
DESCRIPTION:
Host: Department of Statistics\n
Time: October
4, 2012, 3:30pm\n
Location: 300 Seitz Hall\n
Reception:
Following the seminar, please join us
for refreshments at Top of the Stairs (217
College Ave.)\n
\n
Speaker: Netsanet T. Imam, Department
of Biostatistics, State University
of New York at Buffalo\n
\n
Title: Factor Analysis
Regression for Predictive Modeling with High-Dimensional
Data\n
\n
Abstract: We present
factor-model based method to predict a univariate
response, y, as a linear function of explanatory
variables, x = (x1; x2; : : : ; xp),
where the sample size, n, is less than p. We
estimate the coefficient parameters of the model
using bivariate common factor analysis.
We compare the performance of the factor analysis
(FA) regression with partial least squares
(PLS) regression and principal component regression
(PCR) under three underlying correlation
structures: arbitrary correlation, factor
model correlation structure, and when y is
independent of x. Under each structure, we generated
Monte Carlo training samples of sizes
n < p from a multivariate normal distribution
with parameters defining each of three underlying
structures: arbitrary co-\n
variance matrix,
FA covariance structure, and independence
structure. Parameters were fixed at estimates
obtained from analysis of a real datasets,
assuming the parameter restrictions of the
respective structures. Given the independence
structure, we observe severe overfitting by
PLS regression compared to FA regression and
PCR. Under\n
the two dependent structures, FA
regression has comparable average mean square
error of prediction than PCR and PLS regression.\n
\n\n
Price: Free\n
Sponsor: Statistics Department\n
Homepage: http://www.stat.vt.edu/\n
Contact: Christina Dillon\n
Phone: 231-5630\n
E-Mail:: chconne1@vt.edu\n
for more info visit the web at:\n
http://www.stat.vt.edu/Colloquium/colloquia.html\n
LOCATION:300 Seitz Hall
END:VEVENT
END:VCALENDAR