Small-scale VARs have come to be widely used in macroeconomics, for purposes ranging from forecasting output, prices, and interest rates to modeling expectations formation in theoretical models. However, a body of recent work suggests such VAR models may be prone to instabilities. In the face of such instabilities, a variety of estimation or forecasting methods might be used to improve the accuracy of forecasts from a VAR. These methods include using different approaches to lag selection, observation windows for estimation, (over-) differencing, intercept correction, stochastically time--varying parameters, break dating, discounted least squares, Bayesian shrinkage, detrending of inflation and interest rates, and model averaging. Focusing on simple models of U.S. output, prices, and interest rates, this paper compares the effectiveness of such methods. Our goal is to identify those approaches that, in real time, yield the most accurate forecasts of these variables. We use forecasts from simple univariate time series models, the Survey of Professional Forecasters and the Federal Reserve Board's Greenbook as benchmarks