Katharine G. Abraham (University of Maryland), Ron Jarmin (U.S. Census Bureau), Brian Moyer (U.S. Bureau of Economic Analysis) and Matthew Shapiro (University of Michigan) invite you to submit papers to be presented at a conference on the opportunities and challenges associated with the increasing availability of nontraditional sources of data and rapid advances in machine learning and other computational techniques for the production of economic statistics.
The conference will provide economists, statisticians, computer and data scientists from government, academia, business, and non-profit organizations the opportunity to discuss and explore these opportunities and challenges. Authors selected to present papers will be invited to a pre-conference to be held in Cambridge, Massachusetts on July 18, 2018. The conference will be held in Washington, D.C., on March 15-16, 2019.
Location: Washington D.C.
Date: March 15-16, 2019
About the Sponsor: The Conference on Research in Income and Wealth (CRIW) was founded in 1936 to advance the cause of measurement in economics. CRIW brings together economists from government, academia, business, and non-profit organizations to discuss problems of mutual interest.
Background: The coming decades will witness significant changes in the production of the social and economic statistics on which government officials, business decision makers and private citizens rely. Today, the statistical information produced by the federal statistical agencies rests primarily on “designed data”—that is, data collected through household and business surveys. The increasing cost of fielding these surveys, the difficulty of obtaining survey responses and questions about the reliability of some of the information collected have raised questions about the sustainability of that model. At the same time, the potential for using “big data”—very large data sets built to meet governments’ and businesses’ administrative and operational needs rather than statistical purposes—in the production of official statistics has grown.
These naturally occurring data include not only administrative data maintained by government agencies but also scanner data, data scraped from the Web, credit card company records, data maintained by payroll providers, medical records, insurance company records, sensor data, the Internet of Things and other data sources. Provided that the challenges associated with their use can be satisfactorily resolved, these emerging sorts of data could allow the statistical agencies not only to supplement or replace survey data on which they currently depend for producing existing statistics, but also to introduce new statistics that are more granular, more up-to-date and of higher quality than those currently being produced.
Purpose: The purpose of this conference is to provide a forum where economists, data providers and data analysts can meet to present research on the use of big data in the production of federal social and economic statistics. This might involve:
Methods for combining multiple data sources, whether they be carefully designed surveys or experiments, large government administrative datasets or private sector big data, to produce economic and social statistics.
Case studies illustrating how big data can be used to improve or replace existing statistical data series or create new statistical data series.
Best practices for characterizing the quality of big data sources and blended estimates constructed using data from multiple sources.
Presenting authors will have their papers formally discussed by participants drawn from academia, government, business and non-academic research institutions.
Paper submission topics might include:
- Methods for combining and curating data from many sources to improve existing statistics and to compute new high quality statistics from blended data
- Editing and imputation of variables on survey records
- Record linkage to augment the variables available on survey records
- Creation of multiple-frame statistical estimates
- Production of small-domain and other modeled estimates
- Production of high frequency estimates
- Case studies examining how administrative data might be used to inform, augment or replace existing statistical series, such as:
- Using tax and/or program participation data to produce statistics on employment arrangements and the sources of household income
- Using Medicaid and Medicare data to produce statistics about health care providers
- Using administrative data to produce modeled small-domain estimates
- Case studies examining how private data sources might be used to inform, augment or replace existing statistical series, such as:
- Using online or scanner data to produce better measures of prices and/or consumption
- Using credit card and banking data to improve estimates of incomes, spending and trade flows
- Using medical records to improve estimates of medical care spending and health outcomes
- Using satellite image data to improve estimates of economic activity
- Quality metrics for statistics based on blended data:
- Assessing the coverage of statistics based on blended data
- Assessing the quality of record linkages involving blended data
- Assessing the measurement error properties of blended data
Abstracts are invited for submission to the conference organizers by March 19, 2018. Please upload your extended abstract to:
The extended abstract should explain clearly what you will do and (if appropriate) what data you plan to use. Decisions will be made by April 16, 2018. A pre-conference to discuss authors’ plans for their papers will be held July 18, 2018, the Wednesday following the two-day CRIW workshop scheduled for July 16-17, 2018 at the NBER Summer Institute. Final versions of conference papers will be due approximately 1 month prior to the conference date. We anticipate that papers will have discussants.
The papers from the conference will become part of an NBER/CRIW Conference volume to be published by the NBER/University of Chicago Press. All papers will be subject to review by the editors and referees from the NBER and the University of Chicago Press.
The organizers welcome submissions by scholars who are early in their careers, who are not NBER affiliates, and who are from groups that are under-represented in the economics profession.