Metrics to Compare Forecasts

Forecast accuracy is the difference between the actual data and the forecast value. When making a decision we want the most accurate forecasting model possible, but how we measure forecast accuracy depends on how the model will be used. We present three methods of measuring forecast accuracy, mean absolute deviation (MAD), mean squared error (MSE), and mean absolute percent error (MAPE); and discuss when each measure is most appropriate for selecting your forecast model. Interwoven through the lesson is the development of all three metrics in Excel. At the end, we practice model selection using our forecast models developed in Lesson 6. We also introduce Excel's Solver for selecting optimal weights when developing weighted moving average forecast and exponential smoothing models to minimize an error metric.
ericjjesse.wordpress.com/cour...