D -- OPTIMAL INTEGER BOUNDS FOR ENTRIES IN MULTI-DIMENSIONAL AND LINKED STATISTICAL TABLES

Notice Date

August 16, 2001

Contracting Office

DHHS, Program Support Center, Division of Acquisition Management, ADP Branch, Room 5-95 Parklawn Bldg., 5600 Fishers Lane, Rockville, MD 20857

ZIP Code

20857

Point of Contact

Marc L. Long, Contracting Officer, (301) 443-7086

E-Mail Address

Program Support Center (mlong@psc.gov)

Description

The Division of Acquisition Management has a requirement for the National Center for Health Statistics (NCHS), Centers for Disease Control and Prevention (CDC) to provide expert research collaboration and to develop research software tools on the topic of optimal bounding of integer entries in multi-dimensional and linked statistical tables, to test findings, and to publish results. National statistical offices are moving from standard release of data in the format of aggregates presented in two-dimensional tables to release of multi-dimensional tables and linked tables, and also towards providing user-specified tables through public use statistical data base query systems. The National Center for Health Statistics, and particularly its Research Data Center, expects to be at the forefront of these developments. Entries in these tables will be derived from information collected by NCHS on individual persons or establishments. Section 308(d) of the Public Health Service Act requires NCHS to preserve the confidentiality of person and establishment information it collects. For tables of counts (so-called "frequency" or "contingency" tables), confidentiality protection amounts to assuring that the values of small counts cannot be accurately inferred. A basic, but extremely difficult, problem for statistical agencies arising in this context is that of estimating the values of integer entries that have been removed from multi-dimensional or linked tables for confidentiality reasons, for the purpose of assuring that no small counts are accurately estimable. This is sometimes referred to as "confidentiality audit," meaning that these methods provide an assured audit that the data product is free of confidentiality problems. The Government knows of only one source, Dr. Stephen F. Roehrig, Pittsburgh, PA. This proposed contract action is for supplies or services for which the Government intends to solicit and negotiate with only one source under the authority of FAR 13.106-1(b). Interested persons may identify, in writing, their interest and capability to respond to the requirement or submit proposals. This notice of intent is not a request for competitive proposals. However, all proposals received within 15 days after date of publication of this synopsis will be considered by the Government. A determination by the Government not to compete this proposed contract based upon responses to this notice is solely within the discretion of the Government. Information received will normally be considered solely for the purpose of determining whether to conduct a competitive procurement. Any requests for solicitation packages will be considered nonresponsive. All submissions are to be sent to Marc L. Long, at the address above, referencing 01C103124. This acquisition is being made utilizing FAR Part 13, Simplified Acquisition Procedures.

Record

Loren Data Corp. 20010820/DSOL011.HTM (W-228 SN50V259)