Uncertainty Modeling in Dose Response / Statistics in Practice (PDF)

A valuable guide to understanding the problem of quantifying
uncertainty in dose response relations for toxic substances

In today's scientific research, there exists the need to address
the topic of uncertainty as it pertains to dose response modeling.
Uncertainty Modeling in Dose Response is the first book of its kind
to implement and compare different methods for quantifying the
uncertainty in the probability of response, as a function of dose.
This volume gathers leading researchers in the field to properly
address the issue while communicating concepts from diverse
viewpoints and incorporating valuable insights. The result is a
collection that reveals the properties, strengths, and weaknesses
that exist in the various approaches to bench test problems.

This book works with four bench test problems that were taken
from real bioassay data for hazardous substances currently under
study by the United States Environmental Protection Agency (EPA).
The use of actual data provides readers with information that is
relevant and representative of the current work being done in the
field. Leading contributors from the toxicology and risk assessment
communities have applied their methods to quantify model
uncertainty in dose response for each case by employing various
approaches, including Benchmark Dose Software methods,
probabilistic inversion with isotonic regression, nonparametric
Bayesian modeling, and Bayesian model averaging. Each chapter is
reviewed and critiqued from three professional points of view: risk
analyst/regulator, statistician/mathematician, and
toxicologist/epidemiologist. In addition, all methodologies are
worked out in detail, allowing readers to replicate these analyses
and gain a thorough understanding of the methods.

Uncertainty Modeling in Dose Response is an excellent book for
courses on risk analysis and biostatistics at the
upper-undergraduate and graduate levels. It also serves as a
valuable reference for risk assessment, toxicology, biostatistics,
and environmental chemistry professionals who wish to expand their
knowledge and expertise in statistical dose response modeling
problems and approaches.