The nexus of financial development and economic growth across major Asian economies: Evidence from bootstrap ARDL testing and machine learning approach

Cheng Feng Wu, Shian Chang Huang, Tsangyao Chang, Chei Chang Chiou, Hsin Pei Hsueh

研究成果: Article

1 引文 斯高帕斯(Scopus)


The growth trajectory of China, Japan, and India shows their highly influential position in the economy in Asia and the world. A recently developed bootstrap autoregressive-distributed lag (ARDL) test with structural breaks is used to test for cointegration and causality across major Asian economies to investigate the relationship between financial development and economic growth during the period 1960–2016. Although a long-run cointegrating relationship between the time series of real GDP and private credit is insufficient to be detected by a more rigorous test and comprehensive inference, the results show the contingency of causality in the short term. For China, a bidirectional causality with positive supply lead and negative demand following. Positive feedback exists in Japan and India, respectively. In machine learning approach, iterative classifier optimizer with adaboost as iterative classifier performs better than other classifiers in prediction of economic status. The results support the government in Japan and India to keep its steps to regard financial development as an instrument to foster economic growth, and economic growth is considered as an engine to promote financial development for sustainability. The government of China may set proper regulations in the financial market. Additionally, the government may monitor the credit performance to state-owned enterprises and investigate the process of financial resource from input to output, in order to improve financial efficiency and thereby, robustly contribute to economic growth. The empirical analysis provides several policy and managerial implications for decision-makers at the macroeconomic and microeconomic levels.

期刊Journal of Computational and Applied Mathematics
出版狀態Published - 2020 七月


All Science Journal Classification (ASJC) codes

  • Computational Mathematics
  • Applied Mathematics