Possibility of providing pseudo-micro data on official statistics

Abstract

The need for pseudo microdata for education and training has been pointed out to promote the use of official statistical microdata. Currently provided pseudo microdata has a problem in that it does not retain the structure well such as the correlation between variables of the original data. In this paper, I analyzed and examined the method of creating pseudo microdata based on synthetic data method that retains the structure of the original microdata as much as possible while satisfying the current legal and institutional restrictions. As a result, by disclosing the results of intermediate aggregation and regression analysis, it was shown that there is a possibility that the retained pseudo microdata can be created and provided.