Goods and Services Tax Revenue Prediction Using Multiple Linear Regression in Papua New Guinea

Abstract This paper presents a numerical study of tax revenue prediction using multiple linear regression in Papua New Guinea and develops a predictive model for PNG’s tax revenue using historical data from 2010 to 2024. The model incorporates key macroeconomic indicators, including GDP growth, inflation, and mineral export values. Preliminary results indicate that while Goods and Services Tax (GST) remain a stable revenue pillar, Corporate Income Tax (CIT) from the resource sector remains highly liable to change rapidly and unpredictably, especially for the worse. The finding suggests that improving Small Business Tax (SBT) compliance could bridge the projected revenue gap. This study provides a framework for the Internal Revenue Commission (IRC) to enhance budgetary accuracy and debt management strategies. Tax Revenue in Papua New Guinea is primarily administrated by the Internal Revenue Commission (IRC) and consists of both direct and indirect taxes. The model mainly considers Goods and Services (GST), which applies a 10% consumption tax to most goods and services supplied in Papua New Guinea. The regression analysis demonstrates that GDP is a statistically significant predicator of GST revenue and confirms the usefulness of multiple linear regression techniques in fiscal forecasting and economic policy analysis.