Domenico Giannone, Michele Lenza, Giorgio Primiceri, 08 February 2018

The availability of large datasets has sparked interest in predictive models with many possible predictors. Using six examples of data from macroeconomics, microeconomics, and finance, this column argues that it is not usually possible to identify sparse models by selecting a handful of predictors from these larger pools. The idea that economic data are informative enough to identify sparse predictive models might be an illusion.


Vox Talks


CEPR Policy Research