This chapter offers a systematic examination of advanced social network modeling techniques, beginning with an in-depth discussion of the logistic model and Quadratic Assignment Procedure (QAP) regression. It proceeds to explore Bernoulli graph models and p* models, elucidating their significance and application within the realm of social network analysis. The chapter explores deeply into the Exponential Random Graph Model (ERGM) for binary data, addressing its probability distribution, interpretation as a Markovian model, and the concept of sufficient statistics. Special attention is devoted to the estimation of ERGMs, featuring comprehensive discussions and numerical demonstrations of the Markov Chain Monte Carlo method and the Metropolis–Hastings algorithm. This exploration is enhanced by an analysis of the goodness-of-fit for dyadic dependence models and the interpretation of marginal effects within ERGM for binary data. Issues of degeneracy are also thoroughly examined. Temporal adaptations of ERGM, such as the Temporal Exponential Random Graph Model (TERGM) and the Separable Temporal Exponential Random Graph Model (STERGM), are introduced, alongside the Generalized Exponential Random Graph Model (GERGM). The chapter ends with a detailed examination of the Stochastic Actor-Oriented Model (SAOM), outlining its key assumptions. Throughout the chapter, complex concepts are explained through detailed manual calculations and step-by-step examples, complemented by demonstrations using S tata, R, and Python. This theoretical exploration is enhanced by practical applications, with two case studies: one investigating a co-patent network using a temporal exponential random graph model, and the other examining a friendship network using a stochastic actor-oriented model.