The session will focus on survey sampling methods and big data for the production and analysis of survey data. BigData users and BigData research community are fast growing bigger and bigger while statisticians at large seem to be growing divided between enthusiastic and worried, when not downright hostile. From a sampling statistician perspective, the session shall try to decipher the multifaceted relationship connecting BigData to statistical and to sampling methodology. The Speaker is Professor Fulvia Mecatti, Dipartimento di Sociologia e Ricerca Sociale, Università Milano Bicocca, Italy. She will present a paper on What is big in big data: a sampling statistician perspective. Data integration and ML based estimation methods cited in the paper as well as the relevance of big data in producing statistical results will be discussed by Professor David Haziza, Department of Mathematics and Statistics, University of Ottawa, Canada and by Dr. Diego Zardetto, Senior Statistician, Data Production and Methods Unit of the World Bank’s Development Data Group.

What is big in big data: a sampling statistician perspective BigData users and BigData research community are fast growing bigger and bigger while statisticians at large seem to be growing divided between enthusiastic and worried, when not downright hostile. Is BigData also a big step ahead to extract information and actual knowledge from data? Is BigData underplaying statistical inference as we know it? Supplanting stats methodology as a low-cost futuristic option? As a (mainly) sampling statistician, in this talk Prof. Fulvia Mecatti shall try to decipher the multifaceted relationship connecting BigData to statistical and to sampling methodology, starting with the still blurry definition of BigData and discussing some very personal considerations and views.

Organiser: Prof. Monica Pratesi

Chair: Prof. Monica Pratesi

Speaker: Prof. Fulvia Mecatti

Discussant: Mr Diego Zardetto

Discussant: Prof. David Haziza