DG Update: January 2002

1. OPTIMAL TABULAR RELEASES. Research is continuing at NISS on this exciting technology. To illustrate the complexities and the potential, we have applied it to a 14-way table from the CPS. The original table has 435,456,000 cells, of which 76,025 are non-zero (the sparsity is .02 of 1%). Assuming that all 1-way marginals were released, NO 2-way marginal would be releasable! (In this case, some non-zero cells can be bounded within ±1.) A 13-way table was derived from the 14-way table by dropping one variable and aggregating others. It contains 2,592,000 cells, with 41,672 non-zero (sparsity = 1.6%). With a risk criterion that no cell bound be narrower than 3, an optimal release (in terms of the number of tables released) contains 351 subtables defined by a frontier of two 7-way and five 6-way tables. With the bound width increased to 5, the optimal release of 239 subtables is define by a frontier of four 6-way and three 5-way marginal tables.

2. PRESENTATIONS At the Confidentiality, Disclosure, and Data Access Conference,Washington, January, 2002:

3. PUBLICATIONS

Duncan, G. T. (2001) "Confidentiality and Statistical Disclosure Limitation." In International Encyclopedia of the Social and Behavioral Sciences, P. Baltes and N. Smelser, eds., pp. 2521-2525.

Duncan, G. T., Fienberg, S. E., Krishnan, R., Padman, R., and Roehrig, S.F. (2001). "Disclosure Limitation Methods and Information Loss for Tabular Data." Confidentiality, Disclosure and Data Access: Theory and Practical Applications for Statistical Agencies. (P. Doyle, J. Lane, J. Theeuwes, and L. Zayatz, eds.), Elsevier, 135-166.

Fienberg, S. E. and Makov, U. E. (2001). "Uniqueness and Disclosure Risk: Urn Models and Simulation." Research on Official Statistics, 4 23-40.

Lee, J., Holloman, C., Karr, A. F., and Sanil, A. P. (2001). "Analysis of Aggregated Data in Survey Sampling with Application to Fertilizer/Pesticide Usage Surveys." Research on Official Statistics, 4 101-116.

Trottini, M. (2001). "A Decision-Theoretic Approach to Data Disclosure Problems." Research on Official Statistics, 4 7-22.

4. ASSOCIATE DIRECTOR SEARCH. NISS is searching for an Associate Director. Nominations and inquiries are welcome. Please see http://www.niss.org/AD_Search.html for details.

Back to DG Project Updates