Evolution of Forecast Disagreement in a Bayesian Learning Model

Abstract

We estimate a Bayesian learning model with heterogeneity aimed at explaining expert forecast disagreement and its evolution over horizons. Disagreement is postulated to have three components due to differences in: i) the initial prior beliefs, ii) the weights attached on priors, and iii) interpreting public information. The fixed-target, multihorizon, cross-country feature of the panel data allows us to estimate the relative importance of each component precisely. The first component explains nearly all to 30% of forecast disagreement as the horizon decreases from 24 months to 1 month. This finding firmly establishes the role of initial prior beliefs in generating expectation stickiness. We find the second component to have barely any effect on the evolution of forecast disagreement among experts. The importance of the third component increases from almost nothing to 70% as the horizon gets shorter via its interaction with the quality of the incoming news. We propose a new test of forecast efficiency in the context of Bayesian information processing and find significant heterogeneity in the nature of inefficiency across horizons and countries. (2006) in Beijing. We thank Jushan Bai, Badi Baltagi, John Jones, Peter Schmidt, Herman Stekler, Victor Zarnowitz, Arnold Zellner, two anonymous referees, and participants in above conferences for helpful comments and suggestions. We alone are responsible for any remaining errors and shortcomings.

Survey-based forecast distributions for Euro Area growth and inflation: ensembles versus histograms

“…1 Starting with Zarnowitz and Lambros (1987), several economic studies have analyzed the relation between disagreement and various notions of forecast uncertainty. As an important contribution to this literature, Lahiri and Sheng (2010) present a factor model which bridges the gap to structural models of expectation formation (e.g., Lahiri and Sheng 2008). By contrast, our results constitute reduced form evidence on the role of forecast disagreement in the ECB-SPF data.…”

Learning and Heterogeneity in GDP and Inflation Forecasts

“…Lahiri and Sheng (2008) find that the second component, i.e. differences in the weights attached by experts to their prior beliefs, to barely have any effect on GDP forecast disagreement, since professional forecasters place very similar weights on their prior beliefs.…”

Forecast performance, disagreement, and heterogeneous signal-to-noise ratios

“…First, one might use the information that the US SPF provides on the occupation of forecasters. In Lahiri and Sheng (2008), the 10 Results are robust against selecting a higher required number of observations. 11 To limit the influence of outlier observations, we use the square root of M SF E i,h as the dependent variable in (4.1).…”

“…An alternative interpretation is that this term represents differences in forecasters' capabilities to filter/interpret publicly available information (Lahiri and Sheng, 2008). Note that in the case of homogeneous signal-to-noise ratios σ 2 η is equal for all forecasters.…”

“…1 In Lahiri and Sheng (2008) and Lahiri and Sheng (2010a), disagreement across forecasters arises due to differences in individual forecasters' prior beliefs, differences in processing new information, and differences in the relative importance that forecasters attach to their priors and the new information. This framework is especially useful to describe situations in which a sequence of forecasts (with shrinking forecast horizon) is made for one particular random variable, such as an annual growth rate.…”