For ultrahigh-dimensional data, variable screening is an important step to reduce the scale of problem, hence, improve estimation accuracy and efficiency. In this paper, we propose a new dependence measure which called log odds ratio statistic be used under sufficient framework. The approach ensures sufficiency selected input features in modeling regression function enhancement existing margina...

Abstract In the information age, network security has gradually become a potentially huge problem, especially in medical field, it is essential to ensure accuracy and safety of images, patient needs be included with minimal change. Combined log-polar transform (LPT) discrete cosine (DCT), novel robust watermarking algorithm for proposed. It realized lossless embedding into images. process featu...

The fundamental purpose of log and log-like transforms for cytometry is to make measured population variabilities as uniform as possible. The long-standing success of the log transform was its ability to stabilize linearly increasing gain-dependent uncertainties and the success of the log-like transforms is that they extend this notion to include zero and negative measurement values. This study...

Surveys of microbial communities (microbiota), typically measured as relative abundance of species, have illustrated the importance of these communities in human health and disease. Yet, statistical artifacts commonly plague the analysis of relative abundance data. Here, we introduce the PhILR transform, which incorporates microbial evolutionary models with the isometric log-ratio transform to ...