Application Of Artificial Neural Network Model To Forecast Growth Rate Of Service Sector’s Value In Ho Chi Minh City

Abstract Ho Chi Minh City has a trade-service sector structure accounting for over 60% and is also strongly affected by domestic and international macroeconomic factors. Accurate forecasting methods support policy planning and economic management in increasingly volatile global and national economies. The Artificial Neural Network (ANN) model has proven to be highly effective in forecasting macroeconomic indicators such as GDP, unemployment rate, inflation, and export revenue due to its ability to process complex data and detect nonlinear relationships. This study presents the analysis and forecasting of the time series data of the quarterly growth rate of trade services in Ho Chi Minh City from 2011 to 2024, using machine learning models including ANN and several econometric models such as SARIMA, Single moving average, Single Exponential Smooth, and Holt-Winter Additive. The results show that the artificial neural network (ANN) model has superior accuracy compared to other models. The study also provides recommendations for applying ANN in urban economic management, contributing to the sustainable growth of Ho Chi Minh City, and expanding the application of ANN in city-level economic forecasting.