딥러닝 시계열 모형을 이용한 당분기 GDP 예측 성능 분석 = Analysis of QoQ GDP Prediction Performance Using Deep Learning Time Series Model

다국어 초록 (Multilingual Abstract)

In this paper, we proposed an algorithm for predicting GDP growth rate using a deep learning time series model spotlighted recently. The proposed algorithm adopts an ensemble deep learning method to ensure stable prediction performance using a large number of economic time series data with low frequency. It also uses a gradual learning method to ensure adaptive performance even in business fluctuations. By demonstrating that the performance could be improved by using economic sector information in learning, the necessity of convergence with domain knowledge was confirmed and the importance of AI operation technology to provide adaptive predictive power was emphasized.
Through performance comparison with traditional machine learning models for the COVID-19 period, we proved that deep learning could be a relatively reasonable predictive tool under rapid economic fluctuations. The deep learning-based adaptive AI algorithm presented in this paper is expected to be developed into a deep learning-based autonomous adaptive economic prediction system through combination with AI operation technology.

국문 초록 (Abstract)

본 논문에서는 최근 각광 받고 있는 딥러닝 시계열 모형을 활용한 GDP 성장률 예측 알고리즘을 제안한다. 제안한 알고리즘은 빈도가 낮은 다수의 경제 시계열 데이터를 이용하여 안정적인 예측 성능을 보장하기 위한 앙상블 딥러닝 방식과, 경기변동 상황에서도 적응적 성능을 보장하기 위한 점진적 학습 방법을 채택한다. 학습에 있어 경제부문 정보를 활용하여 성능이 개선됨을 보임으로써 도메인 지식과의 융합의 필요성을 확인함과 동시에 적응적 예측력을 제공하기 위한 AI 운영화 기술의 중요성을 강조한다. 코로나-19 기간에 대한 전통적 기계학습 모형과의 성능 비교를 통해 급격한 경기변동 하에서 딥러닝이 상대적으로 합리적인 예측 도구가 될 수 있음을 증명한다. 본 논문에서 제시한 딥러닝 기반의 적응적 AI 알고리즘은 AI 운영화 기술을 통해 딥러닝 기반 자율 적응 경제예측 시스템으로 발전할 것을 기대한다.

참고문헌 (Reference)

1 "e-Country Indicators"

2 "Statistics Korea"

3 S. Lee, "Short-term forecasting system using machine learning and mixed cycle model" Bank of Korea 2017

4 "Nowcasting Report - FEDERAL RESERVE BANK of NEW YORK"

5 E. Ghysels, "MIDAS regressions : further results and new directions" 26 : 2007

6 "Key contents and implications of the transition to the main indicator of quarterly GDP growth"

7 최광신, "K-SuperCast: 빅데이터를 이용한 GDP 성장률 적시예측 시스템" 한국데이터정보과학회 30 (30): 723-743, 2019

8 P. Higgins, "GDPNow: A Model for GDP Nowcasting" Federal reserve bank of Atlanta 2014

9 B. Choi, "GDP Prospect by Combination of Forecasts Using Monthly Data" Bank of Korea 2014

10 K. Choi, "GDP Prediction with Mixed Data Sampling Model" Bank of Korea 81-117, 2016

11 H. Lee, "Development of real-time current quarter economic forecast (GDP nowcasting) system using digital new technology" Bank of Korea 2022

12 "Bank of Korea economic statistics system"

13 N. Woloszko, "Adaptive Trees: a new approach to economic forecasting, OECD Economics Department Working Papers 1593" OECD Publishing

14 X. Wu, "A Data-Driven Gross Domestic Product Forecasting Model Based on Multi-Indicator Assessment" 9 : 99495-99503, 2021