我們在具有高維矩陣值協變量數據的邏輯線性模型中考慮貝葉斯估計，特別是在超高維數據中。這項研究的動機是在經典的邏輯式模型中擴展貝葉斯方法。所提出的估計可以應用於在分類方法上，比較多的案例是有關於有無疾病，事件的是否發生，例如張量判別分析以及常見的成像研究、遺傳學等。我們用模擬研究和陶瓷樣品的化學成分數據集來展示所提出的方法。;We consider Bayesian estimation in a logistic linear model with high-dimensional matrixvalued covariate data, especially in ultra-high-dimensional data. The motivation for this study is to develop the Bayesian approach in classical logistic-style models. The proposed estimates can be applied to classification problems, most of which are related to the presence or absence of diseases and the occurrence of events, such as tensor discriminant analysis and common imaging studies, genetics, etc. Simulation studies and a dataset of the chemical composition of a dataset demonstrate the proposed method.
