Title: Logistic regression with network structure
Authors: Guoyu Guan - Northeast Normal University (China) [presenting]
Abstract: As one of the most popular classification methods, logistic regression model has been extensively studied in the past literature. It basically assumes that individual's class label is influenced by a set of predictors. However, with the rapid advance of social network services, social network data are becoming increasingly available. As a result, how to take this additional network structure to improve classification accuracy becomes an important research problem. To this end, we propose a network based logistic regression model taking the network structure into consideration. Four interesting scenarios about link formation of the network structure are discussed under the NLR model. Furthermore, in order to figure out the impact of network structure on classification, asymptotic properties are derived for the prediction rule under different sparsities of network. Lastly, simulation studies are conducted to demonstrate the finite sample performance of the proposed method, and a real Sina Weibo dataset is analyzed for illustration purpose.