Abstract:Logistic regression is key method for modeling the probability of a binary outcome based on a collection of covariates. However, the classical formulation of logistic regression relies on the independent sampling assumption, which is often violated when the outcomes interact through an underlying network structure, such as over a temporal/spatial domain or on a social network. This necessitates the development of models that can simultaneously handle both the network `peer-effect' and the effect of high-dimensional covariates. In this paper, we develop a framework for incorporating such dependencies in a high-dimensional logistic regression model by introducing a quadratic interaction term, as in the Ising model, designed to capture the pairwise interactions from the underlying network. The resulting model can also be viewed as an Ising model, where the node-dependent external fields linearly encode the high-dimensional covariates. We propose a penalized maximum pseudo-likelihood method for estimating the network peer-effect and the effect of the covariates (the regression coefficients), which, in addition to handling the high-dimensionality of the parameters, conveniently avoids the computational intractability of the maximum likelihood approach. Under various standard regularity conditions, we show that the corresponding estimate attains the classical high-dimensional rate of consistency. Our results imply that even under network dependence it is possible to consistently estimate the model parameters at the same rate as in classical (independent) logistic regression, when the true parameter is sparse and the underlying network is not too dense. We also develop an efficient algorithm for computing the estimates and validate our theoretical results in numerical experiments. An application to selecting genes in clustering spatial transcriptomics data is also discussed.

Statistica Sinica Preprint No : SS-2017-0281 Title Logistic Regression with Network Structure

LOGISTIC REGRESSION WITH NETWORK STRUCTURE

A Dynamic Logistic Regression for Network Link Prediction

Structured Sparse Linear Model For Social Trust Prediction

Link prediction via latent space logistic regression model

Network linear discriminant analysis

Network-based Naive Bayes Model for Social Network

Link Prediction-Based Multi-label Classification on Networked Data.

Studying Determinants of Social Network Size:Comparison of different statistical models

Multi-Layer Stochastic Block Interaction Driven by Logistic Regression (MLSBI-LR) for Efficient Link Recommendation in Intra-Layer Linkage Graphs

A connected network-regularized logistic regression model for feature selection

Link Prediction Using Extended Neighborhood Based Local Random Walk in Multilayer Social Networks

Statistica Sinica Preprint No : SS-2017-0135 Title A Naive Least Squares Method for Spatial Autoregression with Covariates

A Sparse Beta Regression Model for Network Analysis

High Dimensional Logistic Regression Under Network Dependence

A Latent Logistic Regression Model with Graph Data

A Latent Moving Average Model for Network Regression

Link Prediction Based on Dynamic Weighted Social Attribute Network.

Supporting Regularized Logistic Regression Privately and Efficiently

Functional Linear Model with Prior Information of Subjects' Network

A Classification Study on the Strength of Social Relationships Based on Social Networks