Can the economy be as predictable as the weather?
-- Forecasting GDP growth with machine learning
There is an old joke that God created economists to make weathermen look good. Economists owe their notoriety to one task they are not very good at: forecasting. Notwithstanding its difficulty, predicting significant changes to the economy is valuable. With better forecasts, capital allocation and risk management can be performed with far greater efficiency. Keep in mind also that the knock on weathermen has lost its bluster because weather forecasting has improved considerably with the help of data and supercomputers. Can economic forecasting be similarly improved with data and algorithms? We think the answer is yes, and this article describes our use of deep neural networks to forecast GDP.
