Activity Number:
|
78
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 12, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Business & Economics Statistics Section*
|
Abstract - #301800 |
Title:
|
Developing Statistical Models of Non-Compliance for Use with Operational Audit Data
|
Author(s):
|
Chih-Chin Ho*+ and Brian Erard
|
Affiliation(s):
|
Internal Revenue Service and B. Erard & Associates
|
Address:
|
1111 Constitution Avenue, Washington, District of Columbia, 20224, U.S.A.
|
Keywords:
|
Random Audit ; Tax Gap Estimation ; Sample Weight ; Bootstrapping ; Calibration ; Cross Validation
|
Abstract:
|
Traditionally, the IRS has relied heavily on the results from random audits for tax gap estimation as well as workload selection. However, like most state tax agencies, the IRS also has a wealth of statistical information from more recent non-random operational audits that is potentially useful for such purposes. This paper will summarize recent efforts to develop appropriate statistical models for extrapolating the results from operational audits to the general population of returns.
|