Walter Piegorsch: UA's Wizard of Odds Does the Math

The BIO5 statistics expert has co-authored a one-stop shop for everything related to calculating uncertainty and probability, offering the most comprehensive statistics resource available online.
From cancer research to disaster relief, from engineering bridges to playing a poker hand, a hidden force we rarely think about is at play in almost every aspect of life: uncertainty and probability. Yet, at the end of the day, virtually every conceivable process is a game of odds.
Therefore, it should come as no surprise that statistics — the science of understanding and predicting the odds — is one of the most important disciplines in science, health care and economics, to name only a few fields.
"Statistics is everywhere. Biomedical research, risk assessment, environmental sciences, ecology, business, finance, engineering, urban planning ... I could go on," said Walter Piegorsch, a professor of mathematics at the University of Arizona. "And don't even talk about psychology, psychiatric testing, educational testing, and how you measure all that."
Piegorsch is one of six statistics experts recruited by science publishing giant John Wiley & Sons to develop a comprehensive online resource that covers the fundamentals and applications of statistics in all fields where it is widely used. The product is the most inclusive authoritative, online reference available in statistics.
Hosted on Wiley Online Library and marketed at the institutional level, Wiley StatsRef is aimed at statisticians, scientists from other disciplines and practitioners.
"Almost every dataset scientists generate has uncertainty in it," said Piegorsch, who also holds professorships in Agricultural-Biosystems Engineering in the UA College of Agriculture and Life Sciences and the UA's Mel and Enid Zuckerman College of Public Health. He also is a member of the BIO5 Institute, the Institute of the Environment andSISTA, the School of Information: Science, Technology and Arts.
"Quantifying the uncertainty that is inherent in a set of data used to understand a physical phenomenon is one of the driving forces behind why we need statistics."
Statistics is designed to analyze data, to extract information about underlying phenomena, and it's becoming increasingly important in the current age of "big data science," Piegorsch said.
"Tons of data are being generated by the massive boosts in technology that the 21st century has brought us, far more than statisticians have been prepared to handle. And more and more, doing science means mining data, digging through mountains of information to see what is going on.
"For example, say a study reveals tiny differences in the way cells respond to a cancer drug, but there is a lot of random variation, so you can't tell whether there is an effect or not," Piegorsch explained. "You have to be able to step back and understand the statistical methods in order to spot small-but-true effects. If the goal is to analyze big data well, you really have to understand statistical concepts. Even if you hire statisticians to help you make sense of the data, you still need to know what they are talking about. If you have any kind of scientific interest, you have to pay attention to those details."
Wiley StatsRef contains more than 6,000 articles — equivalent to more than 30 printed volumes — and more than 30,000 cross-references and related article links. The database constitutes a stand-alone, self-contained work covering all aspects of the subject in a single product. StatsRef is fully searchable and users can browse by article or topic. Piegorsch and his team of editors update the resource regularly.
"It really is for every researcher and user of statistics," Piegorsch said, "anyone who has a question about an issue in data science and statistics. From 'What's a standard deviation?' to 'How do I fit a nonparametric regression model in more than one dimension?'
"The beauty of statistics is that once a piece of data becomes a number, it behaves like a number — regardless of whether we're talking about the risk of terrorist attack faced by a city or how much diesel exhaust I can breathe before my cancer risk goes up," Piegorsch said.  
"I don't know any other single spot in one place where one can find all the statistics information that has been vetted and peer reviewed — every single article —unlike some other, informal online sources."
Because StatsRef never was designed to be put into print, users have the ability to do advanced searching within any field, for specific words, phrases, times and dates. The content is divided into five areas: applications, concepts, history and biography, methods and theory, and journals and societies.
"There is stuff in here that can bring you to the cutting edge if you want to be an expert," Piegorsch said, "or, if you just want to get basic ideas, you'll find it here, too."

Walter Piegorsch sends e-mail)


Mobile Menu