“The Role of Statistics in the Big Data Centric World”
Location: LTS Auditorium, 8080 Greenmead Drive
There is now an abundance of big data available from different sources such as social media tools, mobile applications, sensors and administrative records. These organic data are typically highly unstructured and contain various types of imperfections.
In this talk, I will discuss how different statistical tools such as sampling, record linkage, statistical matching, and small area modeling can be applied in order to extract potentially useful information from big data for different statistical inferences.
One factor that has slowed down research in big data is a constellation of confidentiality laws and policies related to the release of many big data. In this context, I will discuss the advantage of creating synthetic big data for research purposes that are similar, but not identical to the raw big data. Creation of such synthetic big data has a great promise in advancing big data technology at a much faster rate since it will give researchers access to useful synthetic big data without compromising the confidentiality of the real big data.
Partha Lahiri is a professor of survey methodology and mathematics at UMD and an adjunct research professor at the Institute of Social Research, University of Michigan, Ann Arbor.
Before coming to UMD, he was the Milton Mohr Distinguished Professor of Statistics at the University of Nebraska-Lincoln.
His research interests include big data, Bayesian statistics, record linkage, and small-area estimation. Lahiri has served on a number of advisory committees, including the U.S. Census Advisory committee and U.S. National Academy panel.
Over the years, Lahiri has advised various local and international organizations such as the United Nations Development Program, the World Bank, and the Gallup Organization.
He is a Fellow of the American Statistical Association and the Institute of Mathematical Statistics and an elected member of the International Statistical Institute.