Advances in Statistical Network Modeling and Nonlinear Time Series Modeling

Abstract

The thesis is composed of two independent topics: statistical network modeling and nonlinear time series modeling. With the increasing demand of network data analysis, we present two statistical network models and inferences, analyzing both theoretical and real data performance. The first model targets static networks while the second model targets dynamic networks. The time series modeling focuses on sequential, nonlinear and high dimensional prediction problem which arises with the era of big data.