Frontiers in Statistics
(Leader: Dr. George Yanev)
Friday, November 17, 2006
| Title |
Two Centuries of Hypothesis Testing |
| Speaker |
Eric Lehmann
ASA Continuing Education Program
Distinguished Statistician Video Series |
| Time |
3:00-4:00 p.m. |
| Place |
PHY 013 |
Friday, October 27, 2006
| Title |
Origins of Empirical Bayes (Butterflies, Oysters and Stars) |
| Speaker |
Herbert Robbins
ASA Continuing Education Program
Distinguished Statistician Video Series |
| Time |
3:00-4:00 p.m. |
| Place |
PHY 013 |
Friday, October 6, 2006
| Title |
Statistical Environmental Models: Hurricanes. Predicting
the birth of a storm, predicting the intensity of a storm and analysis
of the Saffir-Simpson scale |
| Speaker |
Rebecca Wooten |
| Time |
3:00-4:00 p.m. |
| Place |
PHY 013 |
Abstract
The relationship between statistics and physics looking at the correlation
between wind speed and pressure versus wind speed and temperature play a
significant role in hurricane prediction. Contrary to previous studies, this study
indicates that a drop in pressure is a result of the storm and less a cause. It
shows that temperature is a key indicator that a storm will form in conjunction
with a drop in pressure.
This study demonstrates a model that predicts the wind speed within a storm
with a high degree of accuracy. With the verified model, we can perform surface
response analysis to estimate the conditions under which the wind speed is
maximized. Additional studies introduce a model that predicts the number of
lightning strikes dependent on significantly contributing factors such as
precipitable water, the temperatures within a column of air and the temperature
range. Using extreme value distribution and historical data we can best fit flood
stages, then including profiling estimate return periods.
Friday, September 29, 2006
| Title |
A Multiple Regression Model for the Prediction of Average
Annual Rainfall in Florida |
| Speaker |
George W. Kimber, Jr. |
| Time |
3:00-4:00 p.m. |
| Place |
PHY 013 |
Abstract
Several multiple regression models for the prediction of average annual
rainfall in Florida will be presented and discussed. The nature, limitations, and
potential of the original pilot study will be reviewed. A comparison will be made
with a similar multiple regression model used to predict average annual rainfall
in California. There will be an analysis of the underlying assumptions. The
commands in SAS that generate the relevant procedures will also be demonstrated.
A general discussion of the multiple regression procedures and of their
interpretations will be presented.
Friday, September 22, 2006
| Title |
Characterizations of Probability Distributions via Bivariate
Regression of Record Values |
| Speaker |
George P. Yanev |
| Time |
3:00-4:00 p.m. |
| Place |
PHY 013 |
Abstract
We characterize the exponential distribution in terms of the regression of
a function of a record value with its non-adjacent record values as covariates.
We also consider a more general setting involving monotone transformations.
We present special cases of characterizations involving weighted arithmetic,
geometric, and harmonic means.
Friday, September 8, 2006
| Title |
Parameter Estimation for the Distribution of Number of Children |
| Speaker |
Julie Cholet |
| Time |
3:00-4:00 p.m. |
| Place |
PHY 013 |
Abstract
The number of children born to a family may be modeled by the negative binomial
distribution. However, the frequency class containing zero children is typically
larger than expected under this distribution. There are two categories of
childlessness. One consists of both deliberate choice and biological infertility,
collectively labeled “sterility”. The other is the operation of
chance. The latter category belongs to the negative binomial distribution, while
the former does not. The inflation of the zero class due to presence of
“sterile” families can make parameter estimation of the distribution
difficult. We discuss three approaches to solving this problem.