|
Activity Number:
|
160
- Time Series Methodology: Modern Practices in Seasonal Adjustment and Software
|
|
Type:
|
Topic-Contributed
|
|
Date/Time:
|
Tuesday, August 10, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section for Statistical Programmers and Analysts
|
|
Abstract #317232
|
|
|
Title:
|
Adapting the Seasonal Adjustment of Local Area Unemployment Statistics to the COVID-19 Pandemic
|
|
Author(s):
|
Richard Tiller* and Jennifer Oh and Lizhi Liu
|
|
Companies:
|
Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics
|
|
Keywords:
|
Automatic Outlier adjustment;
AICC;
Parsimonious models
|
|
Abstract:
|
The Covid-19 pandemic delivered an instantaneous shock to the U.S. labor market in March/April 2020. This crisis presents a challenge to seasonal adjustment of labor force data. In this paper we explore various options for seasonally adjusting series during the pandemic using as examples 856 series from the Bureau of Labor Statistics Local Area Unemployment Statistics program. The first issue is how to prevent distortions in seasonal factor estimation from outliers generated by the pandemic. Since we adjust a large number of series, an automated approach is necessary. This is complicated because at the onset of a pandemic there is little data available to estimate its likely duration and dynamics. We explored a number of options in terms of the mix of outlier types allowed with emphasis on a parsimonious model.
|
Authors who are presenting talks have a * after their name.
