Does Big Data Improve Financial Forecasting? The Horizon Effect
| Autor: | Olivier Dessaint, Thierry Foucault, Laurent Frésard |
|---|---|
| Přispěvatelé: | Ecole des Hautes Etudes Commerciales (HEC Paris), HEC Paris Research Paper Series |
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
History
Polymers and Plastics Process (engineering) Computer science Big data Industrial and Manufacturing Engineering Social media 0502 economics and business Econometrics 050207 economics Business and International Management Financial forecasting Measure (data warehouse) 050208 finance Forecasting horizon business.industry JEL: G - Financial Economics/G.G1 - General Financial Markets/G.G1.G17 - Financial Forecasting and Simulation 05 social sciences Forecasts' informativeness JEL: D - Microeconomics/D.D8 - Information Knowledge and Uncertainty/D.D8.D84 - Expectations • Speculations ComputingMilieux_GENERAL JEL: G - Financial Economics/G.G1 - General Financial Markets/G.G1.G14 - Information and Market Efficiency • Event Studies • Insider Trading Financial analysts' forecasts JEL: M - Business Administration and Business Economics • Marketing • Accounting • Personnel Economics/M.M4 - Accounting and Auditing/M.M4.M41 - Accounting [SHS.GESTION]Humanities and Social Sciences/Business administration business |
| Popis: | We study how data abundance affects the informativeness of financial analysts' forecasts at various horizons. Analysts produce forecasts of short-term and long-term earnings and choose how much information to collect about each horizon to minimize their expected forecasting error, net of information acquisition costs. When the cost of obtaining short-term information drops (i.e., more data becomes available), analysts change their information collection strategy in a way that renders their short-term forecasts more informative but that possibly reduces the informativeness of their long-term forecasts. Using a large sample of analysts' forecasts at various horizons and novel measures of their exposure to abundant data (e.g., social media data), we provide empirical support for this prediction, which implies that data abundance can impair the quality of long-term forecasts. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|