This chapter overviews aspects of the class of vector autoregressive (VAR) and vector autoregressive moving average (VARMA) models. We discuss their structure, some properties and methods for parameter estimation, and provide discussion and links to relevant literature. VAR models, in particular, are workhorses of applied time series analysis and forecasting in many areas, especially based on Bayesian analysis. The basic theory of VAR model fitting and analysis extends to classes of time-varying parameter extensions, or TV-VAR models. The latter are particularly prominent in areas such as macroeconomic modeling and forecasting, as well as engineering signal processing. VAR and TV-VAR models are also special cases of more general multivariate DLMs that follow in Chapter 10.