Strategic judgment: its game-theoretic foundations, its econometric elicitation

We provide a new econometric methodology to detect the bias due to strategic interaction induced by subjective learning. This methodology relies on (i) a new definition of coherence based on the Likelihood Principle, specifically designed for economic forecasting; (ii) an environment named “Scoring Structure”, where a Forecast User interacts with a Forecast Producer and Reality. A formal test for
the null hypothesis of linearity in the Structure is introduced. Linearity implies that forecasts are strategically coherent with evaluations and viceversa. The new test has good small-sample properties and behaves consistently with theoretical requirements. Three case studies on the Federal Reserve Bank’s, the Bank of England’s and the Norges Bank’s forecasts support the endemic nature of the strategic judgment in Macroeconomics. The economic interpretation of the results are discussed.