Prediction of Share Price by Using Machine Learning Algorithm: Multiple Linear Regression Model

The purpose of this paper is to predict the Year End Share Price of Australian Mining Companies. This study investigates the relationship between the various financial metrics and the year-end share price of companies experiencing financial distress. Key metrics such as ‘ BV per Share (${\$}$)’, ‘Enterprise Value (${\$}$)’, ‘Gross CF per Share (${\$}$)’, ‘Cash per Share (${\$}$)’, ‘Sales per Share (${\$}$)’, ‘ROE’, ‘Financial Leverage’, ‘Asset Turnover’, ‘ROA’ are analyzed. The dataset includes [395rows $\times$ 33columns] observations, [Mean Squared Error: 0.38, R-square: 0.90] in Linear Regression and [Mean Squared Error: 0.34, R-square: 0.91] in XG Boost the model achieved an R-squared value of 0.91, indicating high predictive accuracy. The results highlight XG Boost’s efficiency in precisely predicting share prices and have implications for financial decision-making and investment strategies in unpredictable markets.