Working Paper: NBER ID: w27406
Authors: Francesco Bianchi; Sydney C. Ludvigson; Sai Ma
Abstract: This paper combines a data rich environment with a machine learning algorithm to provide new estimates of time-varying systematic expectational errors ("belief distortions") embedded in survey responses. We find that distortions are large even for professional forecasters, with all respondent-types over-weighting their own beliefs relative to publicly available information. Forecasts of inflation and GDP growth oscillate between optimism and pessimism by large margins, with biases in expectations evolving dynamically in response to cyclical shocks. The results suggest that artificial intelligence algorithms can be productively deployed to correct errors in human judgement and improve predictive accuracy.
Keywords: belief distortions; macroeconomic fluctuations; expectational errors; machine learning; economic forecasting
JEL Codes: E03; E17; E7
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| systematic expectational errors in survey responses (C83) | overweighing personal beliefs (D91) |
| systematic expectational errors in survey responses (C83) | underweighting publicly available information (D80) |
| misweighting of economic information (D80) | inaccuracies in predictions of inflation and GDP growth (E31) |
| belief distortions (G41) | inaccuracies in predictions of inflation and GDP growth (E31) |
| belief distortions (G41) | correlation with macroeconomic fluctuations (E32) |
| machine-generated forecasts (C53) | lower mean squared forecast errors than survey respondents (C53) |
| survey forecasts (G17) | systematic errors (C83) |
| belief distortions (G41) | macroeconomic outcomes (E66) |