Modeling Sri Lankan gdp using macroeconomic indicators: an approach using principal component analysis

Abstract

Economics is conventionally divided into two parts, namely, microeconomics and macroeconomics. While microeconomics delves into individual and business decisions, macroeconomics examines the broader decisions made at the county and government levels, providing a comprehensive understanding of the economy as a whole. The macroeconomic indicators are crucial reflectors of the country’s economic status as they underscore their pivotal role in sustaining economic growth. This study focuses on analyzing the relationship between macroeconomic indicators and the economic growth of Sri Lanka. Nineteen macroeconomic indicators were extracted from the CBSL reports and the data were collected for the period of 1976-2018 from the World Bank website. The choice of PCA is strategic due to the pronounced high correlation among the variables. Subsequently, forward regression analysis is conducted to model relationships with identified principal components, aiming to determine the most influential macroeconomic indicators impacting GDP and to identify the most reliable model with the highest predictive power for GDP. The two principal components extracted from the analysis are found to closely mirror government activities and human capital involvement in the economy. The robust predictive power of these two principal components in forecasting GDP is evident, with an impressive R-squared value of 99.74%. This underscores their reliability and effectiveness in predicting economic growth.