Welcome to the Chesapeake Chapter Web Page!



EXECUTIVE COMMITTEE


President
Eric Snyder
USARL
(410)278-0260
esnyder@arl.army.mil

Vice President
Brian Hall
USAEC
(410)306-1026
brian.hall@us.army.mil

Secretary
Grace Deng
USAEC
(410)306-0498
shaofen.deng@us.army.mil

Treasurer
Nancy Dunn
USAEC
(410)306-0454
nancy.dunn@atec.army.mil

Program Chair
Terry Holdren
USARL
(410)278-3376
terry.holdren@arl.army.mil

Chapter Representative
David W.Webb
USARL
(410)278-7014
webb@arl.army.mil

Membership Chair
Jamie Edwards
USARL
(410)278-2467


Newsletter Editor
Brian Hall
USAEC
(410)306-1026
brian.hall@us.army.mil

Statistical Literacy Chair
Terry Holdren
USARL
(410)278-3376
terry.holdren@arl.army.mil

WEB MASTER
Gene Hilkemeyer
IP Plus Consulting Inc.
H. (410) 838-4339
VDLHD (410) 278-2917/4231
hilkemfamily@earthlink.net
eugene.hilkemeyer@us.army.mil


ASA MEETING - May 2009


Greetings!

The Chesapeake ASA-CC program Luncheon Meeting for May will be on Wednesday, 27 May 2009,in the Old Baltimore Room of the Top of the Bay Restaurant on Aberdeen Proving Ground. The pay-as-you-go lunch (no brown bags please); begins at 11:30am, and the talk begins at noon. Our speaker is Dr. Dwayne Nuzman. He will present "An Accumulate-toward-the-Mode Approach to Confidence Intervals and Hypothesis Tests and an Application to Bionomially Distributed Data". Abstract is as the following:


Analyses of binomially distributed data usually depend on the normal approximation to the binomial or, for small sample sizes, the binomial cumulative distribution function. The first approach is only good for an unspecified "sufficiently large" sample size and does not reflect the asymmetry of the bionomial distribution for probabilites other than 0.5. The second approach leads to suboptimal designs. Neither approach extends easily to handle multivariate binomial data. This paper describes a new accumulate to the mode approach to hypothesis tests and confidence intervals and the application of this approach to binomially distributed data. The usual cumulative distribution function, which accumulates probability from areas of lower probability to areas of higher probability, is replaced by an accumulate-toward-the-mode distribution function which accumulates probability from areas of lower probability to areas of higher probability. This approach, when applied to asymmetrically distributed data, produces more powerful hypothesis tests and more accurate interval estimates. This in turn brings better information to decision makers from the evaluation of modeling and simulation as well as test results. In addition, this approach easily extends to analysis of multivariate distributions, providing decision makers with better information on the relative performance of alternatives. One example of this is a recently completed survivability study, which used the Integrated Casualty Estimation Model to examine the vulnerability of soldiers with various levels of protection to fragments and the accumulate-toward-the-mode methodology for analysis. This briefing highlights the mathematical method as well as the study results. Results of the study have been provide to the Natick Soldier Center in support of their evaluation of a proposed change to the currently fielded helmet.









Directions from off post:
Off-post visitors can obtain a pass at the Maryland Gate, Route 715 entrance, by showing a photo ID such as a valid driver's license.



Hope to see everyone there!