Virtual Roundtable Discussions

Time Series and Seasonal Adjustment Estimation During the COVID-19 Pandemic moderated by William R. Bell (U.S. Census Bureau)
March 12,

The pandemic resulting from the SARS-CoV-2 coronavirus illness (COVID-19) has caused major disruptions to business activity, creating ongoing challenges for statistical organizations who measure that activity. Formerly predictable behavior suddenly changed, for example, supermarket sales were up dramatically, and restaurant activity was down. Whether long-term time series patterns have changed remains to be seen, and reviewers must decide how to model the behavior in the midst of uncertainty.

Representatives from statistical agencies will discuss time series and seasonal adjustment problems they've faced due to the pandemic, the approaches they've taken to maintain their official time series outputs in real time, and assessments of their status a year into the pandemic.

View Now: Time Series and Seasonal Adjustment Challenges During the COVID-19 Pandemic Discussion (Roundtable)

Census Bureau
Bureau of Labor Statistics Introduction
Bureau of Labor Statistics
Bureau of Transportation Statistics
Federal Reserve Board

Estimation with Nonprobability Samples moderated by Michael Yan
October 15, 2020

In response to increasing challenges associated with traditional probability sampling, survey organizations have started using nonprobability samples to supplement probability samples in order to improve cost efficiency and timeliness of data dissemination. Researchers and practitioners have proposed three general approaches to estimation from nonprobability samples: quasi-randomization, superpopulation modeling, and doubly robust (e.g., Valliant, 2019). There have been case studies and simulations to evaluate various estimation methods under each of these approaches in terms of bias reduction, confidence interval coverage, and total survey errors (e.g., Ganesh et al., 2017; Yang, et al. 2018, 2019). This roundtable provided a broad review of the field and offered an opportunity for interested researchers to share their experiences. Discussions focussed on successes, challenges, client communications, and future research agenda.

View Now:  GSS Nonprobability Sample Discussion (Roundtable)