Online Program Home
My Program

Abstract Details

Activity Number: 675 - Practical Aspects of Survey Design and Analysis
Type: Contributed
Date/Time: Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
Sponsor: Survey Research Methods Section
Abstract #329780 Presentation
Title: Ad-Hoc Calibration for Rounding Rules with Nonlinear Benchmarks
Author(s): Luca Sartore* and Nathan Cruze and Habtamu Benecha and Andreea Erciulescu and Kelly Toppin and Clifford Spiegelman
Companies: National Institute of Statistical Sciences and USDA National Agricultural Statistics Service and USDA National Agricultural Statistics Service and National Institute of Statistical Sciences and National Agricultural Statistics Service and Texas A&M University
Keywords: Multi-objective discrete optimization; Consistency; Survey estimates; Nonlinear constraints; Combinatorial algorithms; Relative error
Abstract:

Survey reports often provide numerous estimates of population parameters at different levels of aggregation. Benchmarking or calibration adjustments are often applied to NASS estimates of agricultural quantities so that lower-level values aggregate at upper-level values. Prior to publication, the adjusted values need to be rounded according to predefined rules, while benchmarking constraints are satisfied by maintaining the calibration equations and identities on multiple quantities. This paper discusses a method on rounding and adjusting estimates so that nonlinear benchmarks are satisfied. The proposed ad-hoc calibration process is based on a discrete minimization of the L1-norm relative error. The implementation of the proposed algorithm exploits combinatorial techniques to reduce the computational burden. An illustration of the method to NASS survey data is presented.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program