Official microdata related Project
Official data, previously available only in aggregate form, can now be used in the form of raw data (microdata) although this depends on research objectives. This group is involved in research using official microdata and other similar microdata and developing technologies related to their use.
In addition, to enable many researchers to make effective use of official microdata, we are working to improve data accessibility through dissemination and educational activities by the Research Consortium for Official Micro-Data of Japan and by operating on-site facilities for official microdata.
The following subprojects are being promoted by the Center’s members.
In order to securely publish various statistics derived from national questionnaire information, we have developed and proposed suppression techniques to prevent the disclosure of confidential information of survey participants.
Project for the Promotion of Secondary Use System of Official Statistical Data (see here for details)
This project pursues the development and provision of databases and data use environments that contribute to research use. In addition, it supports the MIC-led initiative to develop remote access on-site use facilities nationwide and studies models for their maintenance and dissemination.
This study estimated the expected loss using big data that has been integrated from five regional banks. This includes information on collateral, guarantees, and debt collection, which is not publicly available. Based on this data, the study develops and proposes a versatile estimation method that can be applied to all banks.
This study utilizes microdata on official statistics and conducts empirical research on Evidence-Based Policy Making using causal inference methods with natural experiments. In concrete terms, it measures the causal effects of childcare support policies and welfare policies for the elderly on the labor market. Based on this, it proposes potential mechanisms and disseminates evidence-based improvements to institutional design.