THE BANFF SYSTEM FOR AUTOMATED EDITING AND IMPUTATION
暂无分享,去创建一个
Banff is a generalized system recently developed at Statistics Canada for the automated editing and imputation of quantitative survey data. It evolved from the Generalized Edit and Imputation System (GEIS), which has been used at Statistics Canada since the late 1980s. The system is a collection of nine independent, specialized SAS procedures that provide functionality similar to GEIS. However, Banff is much more flexible with respect to the operating environment and easeofuse. This paper briefly reviews the initial development of Banff, and describes each of the system functions in detail. An overview of the methodology behind the functions is presented. Finally, there is a discussion about the current and future development of the system which is targeted at making Banff even more versatile, including the development of new methodologies and a graphical interface.
[1] D. Holt,et al. A Systematic Approach to Automatic Edit and Imputation , 1976 .
[2] N. Chernikova. Algorithm for finding a general formula for the non-negative solutions of a system of linear equations , 1964 .
[3] Jon Louis Bentley,et al. An Algorithm for Finding Best Matches in Logarithmic Expected Time , 1977, TOMS.