We introduce "smooth diagnosticity." Under smooth diagnosticity, agents over-react to new information defined as the difference between the current information set and a previous information set. Since new information typically changes not just the conditional mean, but also the conditional...

We measure the nature and severity of a variety of belief distortions in market reactions to hundreds of economic news events using a new methodology that synthesizes estimation of a structural asset pricing model with algorithmic machine learning to quantify bias. We estimate that investors...