摘要：Information diffusion refers to the process about passing certain information from one subject to another. It is a typical and critical phenomenon observed in large scale social networks. To statistically model such a phenomenon, a network diffusion model is proposed and studied. The diffusion process is then investigated under the modeling framework from both the short term and long term perspectives. To estimate the model, a maximum likelihood estimator and a moment estimator are proposed, whose asymptotic properties are further established. The resulting estimators are manifested to have a reliable finite sample performance through a number of numerical studies. Lastly, the diffusion of earthquake news on Sina Weibo is analyzed to illustrate the practical usefulness.
关键词：diffusion process; maximum likelihood estimator; moment estimator; social network
