A Macroeconomic Nowcasting Model on an Automated Platform for China

Philip Arestis ； Daniel L. Jia ； Mengshi Zhang

《Frontiers in Economics and Management》 2卷9期 (2021/09) Pp. 1-14

https://doi.org/10.6981/FEM.202109_2(9).0001

摘要

The timeliness and accuracy of macroeconomic monitoring and forecasting is key to the success of the monetary policy. Nowcasting models based on the Principal Component Analysis (PCA) framework and filtering technology have been developed by central banks to make the real-time analysis of the macroeconomic conditions. In this paper, we build a novel nowcasting macroeconomic model for China and utilize the payment data and other economic series to nowcast the economic movements. Additionally, we develop a supervised learning automation platform for this model. The results indicate that our model could be used by the central bank to monitor the trajectory of macroeconomic growth in real-time, providing valuable information for the monetary authority's policy-making activities.

關鍵字

Nowcasting ； Principal Component Analysis ； Supervised Learning Automation

參考文獻

Aastveit A, Trovik T. Nowcasting Norwegian GDP: The Role of Asset Prices in A Small Open Economy [J]. Empirical Economics, 2012, 42(1):95-119.

Bai J, Ng S. Determining the Number of Factors in Approximate Factor Models [J]. Econometrica, 2002, 70(1):191-221.

Banerjee A, Marcellino M, Masten I, et al. Leading Indicators for Euro-area Inflation and GDP Growth [J]. Oxford Bulletin of Economics and Statistics, 2005, 67(67): 785-813.

Bikker J A, Kennedy N O. Composite leading indicators of underlying inflation for seven EU countries [J]. Journal of Forecasting, 1999, 18(4):225-258.

Bok B, Caratelli D, Giannone D. Macroeconomic Nowcasting and Forecasting with Big Data [J]. Annual Review of Economics, 2018, 10(1).