Results of Evaluation of AGGIES for ACES
暂无分享,去创建一个
The U. S. Census Bureau’s Annual Capital Expenditures Survey (ACES) collects data about domestic capital expenditures in non-farm businesses operating within the United States. Analysts manually edit the ACES data using a specified set of editing rules. Although individual edits are straightforward, the hierarchical combination of edits are complicated with several nested levels of simultaneous balance requirements. We investigate the feasibility of replacing the current ACES editing procedures with an automated system based on National Agricultural Statistics Service's generalized edit and imputation system (AGGIES). The AGGIES system solves simultaneous linear-inequality edits using Chernikova-type algorithms for determining the minimum number of fields to change so that a record satisfies all edits. These algorithms can simultaneously deal with a large number of mathematical constraints and have been successfully applied in Statistics Canada's Generalized Edit and Imputation System and Statistics Netherlands' CherryPI system.
[1] N. Chernikova. Algorithm for finding a general formula for the non-negative solutions of a system of linear equations , 1964 .
[2] D. Holt,et al. A Systematic Approach to Automatic Edit and Imputation , 1976 .
[3] David S. Rubin. Technical Note - Vertex Generation and Cardinality Constrained Linear Programs , 1975, Oper. Res..
[4] Todd A. Todaro. Evaluation of the Aggies Automated Edit and Imputation System , 1999 .