Economic Predictions With Big Data: The Illusion of Sparsity

• 成果类型: Article
• 署名作者: Giannone, Domenico; Lenza, Michele; Primiceri, Giorgio E.
• 署名单位: Amazon.com; Centre for Economic Policy Research - UK; European Central Bank; Northwestern University; National Bureau of Economic Research
• 刊物名称: ECONOMETRICA
• ISSN/ISSBN: 0012-9682
• DOI: 10.3982/ECTA17842
• 发表日期: 2021
• 页码: 2409-2437
• 关键词: stock return predictability; variable selection; model-selection; linear-regression; cross-section; time-series; inference; GROWTH; shrinkage; RECOVERY

摘要

We compare sparse and dense representations of predictive models in macroeconomics, microeconomics, and finance. To deal with a large number of possible predictors, we specify a prior that allows for both variable selection and shrinkage. The posterior distribution does not typically concentrate on a single sparse model, but on a wide set of models that often include many predictors.