(Please do not reply directly to this email - instead use the contact
details provided below - [log in to unmask])
We would like to draw your attention to a new, more advanced statistics
course organised by the Statistics Courses Team at the UCL Institute of
Introduction to quantile regression - NEW COURSE given by Dr Marco Geraci
19 October 2011 (Wedn); 10am to 4.30pm; Cost for non UCL/ICH/GOSH affiliated
delegates: £50 (introductory price)
Have you ever wondered whether you can estimate a regression model for the
quantiles (e.g., the median or the 5th percentile) of an outcome without
having to introduce distributional assumptions such as normality? Or whether
you can assess heteroscedasticity (non-constant variance), skewness and
other features of the outcome distribution in relation to a set of
predictors? This introductory course to quantile regression (QR) will enable
participants to estimate and evaluate models that relate one or more
regressors to the quantiles of a response variable. Programming code to
perform QR in Stata and R will be provided. Familiarity with probability and
regression analysis is expected from the delegates, although a brief recap
of the basics will be provided.
More courses and additional dates are available on our website
www.statistics-courses.info <http://www.statistics-courses.info/> .
If you are interested in registering for the course above or others that are
listed on our website or finding out more information about the courses,
please visit www.statistics-courses.info or send us an email on
[log in to unmask] or give us a call on 07730405980.
We look forward to hearing from you.
On behalf of the Statistics Courses Team
Centre for Paediatric Epidemiology and Biostatistics
Institute of Child Health, UCL
30 Guilford Street; London; WC1N 1EH
Telephone for stats courses enquiries: 077 30405 980
Statistics Courses enquiries/registrations: [log in to unmask]
<mailto:[log in to unmask]>
P Please consider the environment before printing this e-mail
You may leave the list at any time by sending the command
to [log in to unmask], leaving the subject line blank.