Taking the temperature : forecasting GDP growth for mainland in China

Abstract

We present a new composite leading indicator of economic activity in mainland China, estimated using a dynamic factor model.Our leading indicator is constructed from three series: exports, a real estate climate index, and the Shanghai Stock Exchange index.These series are found to share a common, unobservable element from which our indicator can be identified.This indicator is then incorporated into out-of-sample one-step-ahead forecasts of Chinese GDP growth.Recursive out-of-sample accuracy tests indicate that the smallscale factor model approach leads to a successful representation of the sample data and provides an appropriate tool for forecasting Chinese business conditions. Keywords: Forecasting, China, Leading Indicator, Factor Model, Growth Cycles JEL-Classification: C32, C52, E32, E37