DMS Statistics and Data Science Seminar

Speaker: Stéphane Guerrier, University of Geneva 

Title: Assessing Coronavirus Disease 2019 Prevalence with Sample Surveys and Census Data with Participation Bias

Abstract: Countries officially record the number of Coronavirus disease 2019 (COVID-19) cases based on medical tests of a subset of the population with unknown participation bias. For prevalence estimation, the official information is typically discarded and, instead, random survey samples are taken. One exception is the surveys recorded by the Statistics Austria Federal Institute in cooperation with the Medical University of Vienna and the Red Cross Austria to study the prevalence of COVID-19 in Austria in April, May and November 2020. The survey contains information about the number of positive COVID-19 tests in the sample as well as the participants with a positive COVID-19 test measured through the official procedure in the population during the same period. In this paper, we propose to combine both information, i.e. the outcomes from the survey sample and the census with participation bias, to analyse the prevalence of COVID-19 in Austria in 2020. We show why combining both sources of information improves the efficiency of the prevalence estimator, especially when the prevalence and the survey sample sizes are relatively small. Put differently, using the proposed method, the same level of precision can be obtained with substantially smaller survey sample sizes. We also adjust the estimation method for measurement errors due to the sensitivity and specificity of the medical testing procedure and to the nonrandom sample weighting scheme of the survey.

Join work with Christoph Kuzmics (Department of Economics, University of Graz, Austria) and Maria-Pia Victoria-Feser (Geneva School of Economics and Management, University of Geneva, Switzerland).