U.S. flag

An official website of the United States government

Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Bureau of Labor Statistics > Office of Survey Methods Research > Publications > Browse Research Papers

Predicting the Effect of Business Births and Deaths on the Current Employment Statistics Survey: Using Sample Information to Minimize Coverage Error

Chris Grieves, Steve Mance, and Collin Witt

Abstract

The Current Employment Statistics (CES) program uses a model to account for the bias in monthly payroll employment estimates arising from establishment births and deaths that fall outside the survey frame. Actual birth-death values derived from administrative counts are available with a substantial lag but must be predicted for the current month's estimation. The existing CES net birth-death model relies on the bias from business births and deaths to follow a consistent, seasonal pattern, characterized by an ARIMA process. This has broken down during extreme changes in the labor market necessitating interventions in the model during the COVID-19 recession and recovery. Previous research showed the ARIMA models can be improved by including covariates available coincident with the survey. This paper explores several modelling frameworks that use information from the CES survey to predict birth-death values, substantially outperforming previously examined models over the period from 2007–2020, covering the Great Recession through the pandemic. Forecast combination techniques are also examined and compared with predictions from the individual frameworks.