Bayesian multiple linear regression model for GDP in India

Gross Domestic Product (GDP) known as the pulse of economy for any country depends on multiple factors like export–import, inflation rate and unemployment rate etc. Statistical assessment of GDP demands fresh concepts to explain GDP through its covariates in order to improve and strengthen the estimation process. In the present paper, a linear regression model is proposed for modelling GDP of India. Descriptive analysis for the considered data on GDP and its covariates obtained from the World Bank archives is presented through Heatmap. Identification and relevance of the possible set of covariates is established by using step-wise regression (SR). The use of SR reflects its credibility vis-à-vis Ordinary Least Squares (OLS) regression by registering a decline in Deviance Information Criterion (DIC) from − 18.30 to − 13.63, from a full model to only a few significant factor models. We propose an alternative statistical algorithm implemented under Bayesian paradigm through Integrated Nested Laplace Approximation which bridges the gap of accuracy in estimates as opposed to the frequentist OLS regression for explaining GDP of the country India. Model selection is decided from a battery of normal priors through DIC. Comparison of Bayesian and frequentist modelling results is done using several criteria such as Mean Absolute Deviation, Root Mean Square Error, and Mean Absolute Percent Error. India is one of the emerging economies, the economy of India is the 5th largest in terms of nominal gross domestic product in the world, and the 3rd largest in the terms of purchasing power parity as recorded by World Bank (2022). GDP of India for 2022 is recorded as $3.38 trillion World Bank (2022) which is in close agreement to GDP of $3.361 trillion predicted by the model proposed in the present paper.