Time Series Analysis, Analysis of Correlated Categorical Data, Environmental
Statistics, General Methodology, Data Mining and Analysis.
My research interest is in studying models for correlated data. My published
work includes theoretical developments in nonstationary autoregressive
models, detection of abrupt changes in the parameters of time series models,
and developing overdispersion models to describe categorical data collected
in clusters. I have been involved in a number of environmental applications
of statistical methodology such as trend estimation, ranked set and
composite sampling and investigation of data quality issues using data
mining. I am also involved in analysis of transportation data.
My work in the change point detection problems are relevant to a variety
of applications including financial, economic and environmental time series
data. My work in overdispersion models are applied epidemiological and
teratological data where the basic unit of sampling is a cluster (or a
litter) rather than an individual. These models are also useful in marketing,
telecommunications and other business applications.
Statistics with S-PLUS, my book with Steven P. Millard of MathSoft)
is published by Chapman Hall/CRC Press in 2000. It gives an instant access
to statistical methodology relevant to USEPA policy documents and detailed
examples including S-PLUS implementations of these methods.