An official website of the United States government
United States Department of Labor
United States Department of Labor
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.
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.
Office of Survey Methods Research
SHARE ON:
More Information
Published Year: 2026
Survey/Program(s): Office of Prices and Living Conditions
Catalog: Economic Working Paper
WP Number: WP-589
Download PDF
Implementing a Data Driven Model for Automated Dynamic Outlier Detection During Data Validation
Abstract
When constructing the monthly Producer Price Index (PPI), economists at the Bureau of Labor Statistics (BLS) validate price changes that exceed tolerance levels set within the PPI processing system. Historically, PPI economists used prior industry knowledge to set initial industry tolerance levels when a new sample was introduced. Tolerance levels remained static across processing months unless an economist changed the values based on updated industry information. The BLS recently introduced a machine learning model to improve outlier detection in micro data for the PPI program. This model is now used to estimate and provide PPI economists with monthly-updated, data driven tolerance boundaries to validate price changes. The end goal of this project is to both improve the quality of and create operational efficiencies in outlier detection and micro data review by using a more dynamic method for setting price tolerance levels. This paper explains our approach to this project, the methodology used, the challenges faced, and its implementation into the PPI production process.