Analyzing Nonresponse Bias in PPI Data

• Activity Number: 409
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Government Statistics
• Author(s): Teresa Hesley*+ and Boriana Chopova and Greg Kelly and Soon Paik and Andy Sadler and Dave Slack
• Companies: Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics and Bureau of Labor Statistics
• Keywords: nonresponse bias ; unit initiation ; item repricing ; item short term relative ; sample adjusted weights

Abstract

The Producer Price Index (PPI) program conducted a study to determine if nonresponse bias exists in PPI data. The study investigated nonresponse at unit initiation (when units are asked to participate in the PPI and provide initial item prices) and during the item repricing period (when units provide follow-up prices for the items they have agreed to reprice at initiation). The study consisted of three stages: In the first stage, regression models were used to analyze the relationship between unit initiation response and several explanatory variables as well as between item repricing response and these same set of explanatory variables. In the second stage, another regression model was used to analyze the relationship between item short term relatives and the explanatory variables which were determined to be significant in the first stage. The third stage tested the impact of adjusting s