Judgment can spur long memory

Abstract

We arrive at this conclusion by using a new family of models—the Long Memory Dynamic Judgmental Protocol (LMDJP)—where robust filtering and fractionally integrated auto-regressions are combined in an environment characterized by several players—namely, Forecast Producer, Forecast User, and Reality. Namely, we show that if judgment is parametrized as a deformation Likelihood function according to Lq-Likelihood methods, such a deformation affects (sometimes dramatically) the Power Spectrum, consequently inducing over-rejection in formal tests for no LM-effects based on the last. Our simulated and empirical evidence reveals that knowledge of the fractional integration parameter matters for the p-values of tests for spurious LM and, secondly, that the role of LM in belief formation is ambiguous.