Statistical theory on data protection for public database

研究了用于估计种群规模指数的超总体模型的一些性质。提出了超总体模型之一的皮特曼模型的非参数估计方法。提出了多尺度指标，并将其应用于序列数据、分层样本和分层后样本的微观数据集的风险评估。这些理论和技术被应用到其他领域，如PPDM、空间统计、医学统计和生物统计。

Some properties of the superpopulation models, which are used for the estimation of the population size indecies, are investigated. The nonparametric estimation method associated with the Pitman model, which is one of the superpopulation models, are proposed. The multi size indecies are also proposed and applied to the risk assessment of the microdata set made from the sequential data, the stratified sample and the post-stratified sample. These theories and techniques are adopted to the other fields, for example, PPDM, the spatial statistics, the medical statistics and biostatistics.