[Cowystats] DU/ ASA Seminar Friday March 9 / useR!2007 conference

Matt Pocernich pocernic at rap.ucar.edu
Mon Mar 5 13:25:45 MST 2007

## Mar  9   University of Denver - data mining and geospatial statistics.
## Mar  28  UCHSC Biometrics seminar (noon) 
## Apr  20  CO/ WY Spring meeting
## Apr  27  Seminar Colorado School of Mines (tentative)
## Sept 13, 2007 Colorado Day (SAS Conference), details to follow. 

This note is mainly to provide details on the talks to be held at the 
University of Denver this Friday.  These talks illustrate the
application of a couple statistical techniques to very different
types of problems.  A map to the seminars is attached.  In addition to
being a good reason to leave work early on Friday, this could
be a great opportunity to try the new light rail.

Recently time and location of the first North American useR!
conference has been announced.  (useR! = a graphically clever way
of describing an user of R software. ) This event will be held at Iowa
State, August 8-10th. More information can be found at the R Project
website. (www.r-project.org).


Hosted by the Department of Statistics and Operations Technology
University of Denver and Colorado/ Wyoming Chapter of the ASA 

Room 240
Daniels College of Business
University of Denver

Friday March 9th  2:00 - 4:30

Emma Vazirabadi
Director of Workforce Analytics

Targeting and Saving High Value Employees:  A priceless HR Concept

Apartment Investment and Management Company (AIMCO) is the nation's
largest owner and operator of apartment communities, with nearly 1,370
communities that include approximately 240,000 units.
AIMCO Human Resources with more than 60 employees and six internal
departments is a structured department within in the organization that
strives for recruiting employees from the best employee markets, and
providing competitive compensation and benefits.    

The use of various data mining approaches as it pertains to HR modeling
is multi faceted. However, the SPSS predictive modeling solution,
Clementine, will allow us reduce recruiting costs substantially by
implementing a Candidate Profiling predictive modeling system that uses
our current employee data (demographic, resume data, performance review
data) to profile our best performers and score their data against new
candidate to predict best performers for particular job description.
This essentially empowers the AIMCO Talent team with the ability to only
target and hire the right candidates, and in turn optimize resources,
reduce employee turnover and minimize recruiting costs


Tim Coburn
School of Information Technology and Computing
Abilene Christian University

A local-spatial prediction approach for estimating the volumes of
recoverable unconventional natural gas

This presentation describes an algorithmic approach to estimating the
volume of recoverable, unconventional natural gas that combines features
of cross-validation with the jackknife and bootstrap procedures. The
procedure yields estimated volumes of gas at individual untested
drilling sites as well as regional estimates that can be used in
economic assessments. Simulation experiments using known data from the
Michigan Basin indicate that the estimated total recoverable volume is
within four percent of the observed amount. The algorithm provides a
reasonable alternative to geostatistical approaches which require more
direct knowledge of the spatial trend in the subsurface.

This work is sponsored by the Eastern Energy Resources Division of the
U.S. Geological Survey, which is responsible for producing estimates of
the remaining energy resources used to support the strategic initiatives
of various governmental agencies.

### Keith A. Baggerly from the University of Texas M. D. Anderson
Cancer Center Thurs March 29th: Preventive Medicine and Biometrics
Seminar 12-1pm CPH Auditorium 1J05, 9th St Campus, UCHSC sponsored
with the Bioinformatics Department at the University of Colorado
Health Science Center. (Abstract to follow) 

Matt Pocernich
National Center for Atmospheric Research
Research Applications Laboratory
(303) 497-8312
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ASA Map.doc
Type: application/msword
Size: 460288 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/cowystats/attachments/20070305/3a389908/ASAMap-0001.doc

More information about the Cowystats mailing list