Chicago R User Group (Chicago RUG) Data and Statistics Message Board › Data mining, statistics and R by Johannes Ledolter and Robert McCulloch in M
The Milwaukee Chapter of the ASA presents
The 2014 Annual Meeting
Data mining, statistics and R by Johannes Ledolter and Robert McCulloch
April 4, 2014 at UWM
Current members and students $40
New/former members $49
You can make a reservation by sending a check made out to
the Milwaukee Chapter of the ASA to the Secretary/Treasurer:
Medical College of Wisconsin
Center for Patient Care & Outcomes Research
8701 Watertown Plank Rd.
Milwaukee, WI 53226
ATTN: Elizabeth Smith
Or via Paypal: firstname.lastname@example.org
You will receive an email confirmation and more details.
Please feel free to share this message with others who may be interested.
Johannes Ledolter, PhD
C. Maxwell Stanley Professor of
International Operations Management
Tippie College of Business
University of Iowa
Data mining and business analytics with big and small data
- it is not always size that matters
Abstract: Will review useful methods for data mining
and business analytics describing several applications
and case studies where these methods have proven
useful. Also will discuss the importance of collecting
one's own data through carefully designed statistical
Ledolter, J.: Data Mining and Business Analytics with R.
Wiley & Sons, June 2013. See
for the R programs that are used in the book.
Ledolter, J. and Swersey, A.J.: Testing 1 - 2 - 3:
Experimental Design with Applications in Marketing and
Service Operations, Stanford University Press, 2007.
Robert E. McCulloch, PhD
Katherine Dusak Miller Professor of Econometrics and Statistics
Booth School of Business
University of Chicago
Data Mining with Bayesian Trees
Abstract: Tree regression and classification modeling play a
fundamental role in data mining. In addition to being an important
method in their own right, trees are often used in ensemble methods
such as bagging and boosting in which high dimensional models are
built up from many trees. In this talk, we review practical Bayesian
approaches to data mining with trees and illustrate simple, easy to
use, software implementations which are available in R.
Fitting trees to data presents interesting challenges. There are many
possible tree structures so that searching the tree space for an
appropriate model is challenging. Bayesian inferential methods have
much to offer. The prior specification allows for parsimonious
regularization of the fit as well as nonparametric model
specification. Markov chain Monte Carlo provides a framework for
specifying stochastic searches in this vast model space. Recent
advances in parallel computing have made these Bayesian approaches
practical for large data sets.
09:15-10:30 Robert McCulloch: Part I
11:00-11:30 Robert McCulloch: Part II
11:30-12:00 Johannes Ledolter: Part I
12:00-01:00 Lunch on your own
01:00-02:30 Johannes Ledolter: Part II
03:00-04:30 Johannes Ledolter: Part III
05:00-? Dinner with the speakers: Restaurant TBD; please RSVP
More details can be found on our web page