Moving average and stochastic volatility are two important components for modeling and forecasting macroeconomic and financial time series. The former aims to capture short-run dynamics, whereas the latter allows for volatility clustering and time-varying volatility. We introduce a new class of models that includes both of these useful features. The new modelsallow the conditional mean process to have a state space form. As such, this general framework includes a wide varietyof popular specifications, including the unobserved components and time-varying parameter models. Having a moving average process, however, means that the errors in the measurement equation are no longer serially independent, and estimation becomes more difficult. We develop a posterior simulator that builds upon recent advances in precision-based algorithms for estimating this new class of models. In an empirical application involving U.S. inflation we find that these moving average stochastic volatility models provide better in-sample fitness and out-of-sample forecast performance than the standard variants with only stochastic volatility.