In this paper, we introduce the concept of sparse bilinear logistic regression for decision problems involving explanatory variables that are two-dimensional matrices. Such problems are common in computer vision, brain-computer interfaces, style/content factorization, and parallel factor analysis. The underlying optimization problem is biconvex; we study its solution and develop an efficient al...

Multinomial logistic regression is the extension for the (binary) logistic regression when the categorical dependent outcome has more than two levels. For example, instead of predicting only dead or alive, we may have three groups, namely: dead, lost to follow-up, and alive. In the analysis to follow, a reference group has to be chosen for comparison, the appropriate group would be the alive, i...

Logistic regression with binary and multinomial outcomes is commonly used, and researchers have long searched for an interpretable measure of the strength of a particular logistic model. This article describes the large sample properties of some pseudo-R statistics for assessing the predictive strength of the logistic regression model. We present theoretical results regarding the convergence an...

Logistic regression is an important statistical analysis methods widely used in research fields, including health, business and government. On the other hand preserving data privacy is a crucial aspect in every information system. Many privacy-preserving protocols have been proposed for different statistical techniques, with various data distributions, owners and users. In this paper, we propos...

Logistic regression is a commonly used representation for aggregators in Bayesian belief networks when a child has multiple parents. In this paper we consider extending logistic regression to relational models, where we want to model varying populations and interactions among parents. In this paper, we first examine the representational problems caused by population variation. We show how these...

We advocate Logistic Regression (LR) as an alternative to the Support Vector Machine (SVM) classifiers commonly used in steganalysis. LR offers more information than traditional SVM methods – it estimates class probabilities as well as providing a simple classification – and can be adapted more easily and efficiently for multiclass problems. Like SVM, LR can be kernelised for nonlinear classifi...