The Stock Prices Prediction Model for Construction Companies Using Artificial Neural Network in Vietnam

Abstract:

This study aims to predict the stock prices based on factors affecting the stock prices of construction companies listed on the Ho Chi Minh City Stock Exchange by using Artificial Neural Networks (ANN). The research data is collected from 22 construction companies listed on the HOSE over a period of 29 quarters, with a total of 595 samples. Study primarily uses the Artificial Neural Network (ANN) model and uses Multivariate Linear Regression (MLR) for comparison. The results show that the ANN model performs better, with six significant variables: Earning Per Share (EPS), Book Value Per Share (BVPS), Return on Assets (ROA), Inflation Rate (INF), Interest Rate (INT), and Construction Steel Price (SP). Meanwhile, the linear regression model only identifies three significant variables. This study also ranks the independent variables based on their impact level, in descending order: BVPS, EPS, ROA, Inflation Rate, Interest Rate, and Construction Steel Price.