Measuring Market Uncertainty
Anthony Davies and Kajal Lahiri
In this paper, we use probability forecasts from the ASA-NBER Survey of Professional Forecasters to separate estimates of market uncertainty from forecaster disagreement. We show that measures of disagreement (which have been used in the past as proxy measurements of uncertainty) have a low correlation with uncertainty. We use our estimates to measure the impact of news on market uncertainty (as opposed to market volatility) and find (1) that bad news has a greater impact on uncertainty than good news, (2) that the larger is the magnitude of the news the proportionally larger is the market uncertainty (i.e. the magnitude of news exponentially affects uncertainty), and (3) that uncertainty in one period is not significantly affected by uncertainty in the previous period.
Return to Cline & Davies Research Alliance