South Korea’s central bank said on Sept. 2 it has launched research into adopting artificial intelligence deep learning to extract economic growth forecasts, amid growing uncertainties stemming from the coronavirus pandemic.

The Bank of Korea expressed anticipations of predicting the nation’s gross domestic product readings and other key economic indexes by utilizing deep learning, which is a subset of AI technology, in its regular economic report. The research is still in its early stages, it added. 

A man walks out of the Bank of Korea headquarters in Seoul. (Yonhap)

As part of its research, the BOK compared the results of deep learning-based calculations and traditional econometrics methods to make short-term forecasts on monthly exports and the daily won-dollar exchange rate.

While the outcomes were similar, deep learning produced results with a noticeably narrow margin of error compared with the traditional method, especially when predicting exchange rate forecasts, the BOK noted.

“Compared to conventional econometric approaches such as VECM, the deep learning approach shows not only better prediction powers, but also more informative error bands that may indicate periodical developments of uncertainty in the economy,” it said in the report, published in English.

“In this regard, we can conclude that deep learning is useful, even with data that seems to not be coordinated with any machine learning approach,” it added.

But the BOK also stressed there is room for improvement, saying that the study is no more than a prototype that can be improved with “enhanced data.”

By Jung Min-kyung (mkjung@heraldcorp.com)