How prevalent is computer simulation in macro-economic research and forecasting?

The question revolves around another motivating question: How does economics become more effective at predicting the behavior of the macro economy? For example, how does economics develop methods that allow it to predict with some certainty, more certainty than current methods provide, that a recession in country A will occur? How much is simulation used by economists and institutions? I know that the Federal Reserve in the US conducts bank 'stress tests', which are a sort of simulation. I'm wondering if this approach has been expanded and scaled to answer other sorts of economic questions. Are there any institutions or researchers developing simulations as a core part of their research?