search for




 

K-SuperCast: A big data based GDP forecasting model
Journal of the Korean Data & Information Science Society 2019;30:723-43
Published online July 31, 2019;  https://doi.org/10.7465/jkdi.2019.30.4.723
© 2019 Korean Data and Information Science Society.

Kwang-shin Choi1

Financial Supervisory Service
Correspondence to: Senior researcher, Macroprudential Supervision Department, Financial Supervisory Service, 38 Yoeui BLVD, Seoul 07321, Korea. E-mail: kwangshin.choi@gmail.com
Received May 21, 2019; Revised June 5, 2019; Accepted June 5, 2019.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
This study introduces K-SuperCast, a big data-based GDP growth forecasting system. K-SuperCast strengthens the reliability of forecasts by mitigating the limitations on the type and time range of explanatory variables that the existing GDP growth prediction models based on time series data had. Reliability test results are as follows: K-SuperCast satisfies consistency condition with finite samples and back-testing shows that the deviation between the prediction and the actual value is evenly distributed around zero. These results show that K-SuperCast has practical usability. Finally, results suggest three main factors that determine GDP growth rate and they are [1. Corporate performance and prospect], [2. Construction and real estate] and [3. wages and labor], so we need to closely monitor these factors to supervise the stability of the financial system.
Keywords : GDP Growth , Nowcasting, Factor Model, Kalman Filtering