In traditional tests of forecast rationality, price forecasts are usually differenced to obtain stationarity. However, this data transformation may ignore important long-run information contained in forecasted price levels. Here, the concept of forecast consistency is paired with rationality concepts used in the market efficiency literature to develop a sequential testing procedure for forecast consistency and rationality. USDA quarterly livestock price forecasts do not demonstrate long-run consistency.
