Time Series Analysis of Regional GDP and Stock Market Volatility Using ARIMA and ARMA-GARCH Models

Mingrui Ma

doi:10.54097/4t750n32

Authors

Mingrui Ma

DOI:

https://doi.org/10.54097/4t750n32

Keywords:

ARIMA, GARCH, GDP forecasting, stock volatility, time series analysis.

Abstract

This study applies ARIMA and ARMA-GARCH models to analyze quarterly GDP data (2007-2019) for Beijing, Tianjin, and Hebei, and daily Shanghai Composite Index prices (2002-2021). The ARIMA(1,0,0)(0,1,0) model demonstrates predictive accuracy for regional GDP (MAE = 8.5%), while the ARMA(0,0,3)-GARCH(1,1) model effectively captures the "peak and fat tail" characteristics of stock returns. Findings provide actionable insights for economic policymakers and financial risk managers, highlighting the models' effectiveness in addressing non-stationarity and volatility clustering in economic and financial time series.

Downloads

Download data is not yet available.

References

[1] He Y, Li J M, Ruan S M, et al. A Hybrid Model for Financial Time Series Forecasting-Integration of EWT, ARIMA With the Improved ABC Optimized ELM [JJ. IEEE Access,2020, (8).

[2] Cibicong, Zhang Pinyi. Financial Time Series Prediction Based on ARIMA-LSTM Model [J]. Statistics and Decision, 2022, 38(11): 145-149. DOI: 10.13546/j.cnki.tjyjc.2022.11.029.

[3] Engle, R. F. (1982). Autoregressive conditional heteroskedasticity with estimates of the variance of the United Kingdom inflation. Econometrica, 50(3), 987-1007.

[4] Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 31(5), 307-327.

[5] Taylor, S. J. (1986). Modeling financial time series. Chichester: John Wiley and Sons, pp. 35-40.

[6] Kim H Y, Won C H. Forecasting the Volatility of Stock Price Index: A Hybrid Model Integrating LSTM With Multiple GARCH-type Models[J]. Expert Systems with Applications,2018, (103).