Machine Learning Modelling of GDP Datasets of Botswana

The main objective of this paper is to investigate how exogenous variables X= (X1, X2, …, X14) affect the gross domestic product (GDP) of Botswana data (Y) using different multivariate methods such as PCA, linear discriminatory analysis (LDA), factor analysis and other classifiers. Using supervised learning techniques, machine learning models are fitted and performance is monitored against test data, calculating the error between observed Y and predicted Y in each case. The results show 100% accuracy for some classifiers and more than 75% accuracy for some other classifiers.