A Method of Masking Based on Multiplicative Noise Jeong, Dong-Myeong; Kim, Jay-J.; Kim, Kyung-Mi;
According to the type of microdata, the various methods have been in use for masking microdata. Multiplicative noise is the one of popular schemes for masking continuous variables. In this paper, we introduce the method of masking based on multiplicative noise and show some results of the application on the 2006 Householder Income and Expenditure Survey (HIES) data. To create the multiplicative noise factor, we used the triangular distribution. truncated triangular distribution, trapezoidal distribution, and double triangular distribution. Also, formulas for the domain estimation for the data masked by the multiplicative noise are developed.
Fuller, W. A. (1993). Masking procedures for microdata disclosure limitation, Journal of Official Statistics, 9,383-406
Kim, J. (1986). A method for limiting disclosure in microdata based on random noise and transformation, In Proceedings of the Section on Survey Research Methods, American Statistical Association, 370-374
Kim, J. (2007). Application of the truncated triangular and trapezoidal distributions for developing multiplicative noise, In Proceedings of the Section on Survey Research Methods, American Statistical Association, CD-ROM
Kim, J. and Winkler, W. E. (2001). Multiplicative noise for masking continuous data, In Proceedings of the Section on Survey Research Methods, American Statistical Association, CD-ROM