Application of Principal Component Analysis and Hierarchical Regression Model on Kenya Macroeconomic Indicators

The aim of this paper was to apply Principal Component Analysis (PCA) and hierarchical regression model on Kenyan Macroeconomic variables. study adopted a mixed research design (descriptive correlational designs). 18 macroeconomic variables data were extracted from Kenya National Bureau Statistics World Bank for the period 1970 2019. R software utilized conduct all analysis. used reduce dimensionality data, where original set matrix reduced Eigenvectors Eigenvalues. A fitted components, R2 determine whether components good fit predicting economic growth. results showed that first component explained 73.605 % overall Variance highly correlated with 15 Additionally, second principal described approximately 10.03% total Variance, while two had higher positive loading into it. About 6.22% variance by third component, which only one first, second, models F statistics 2385.689, 1208.99, 920.737, respectively, each p-value 0.0001<5% hence implying significant. lowest mean square error 17.296 as best predictive model. Since 1 highest explained, lower than other models, more reliable in explaining Therefore, it concluded associated monetary economy, trade openness economy government activities, consumption factor investment predict growth Kenya. recommends PCA should be when dealing variables, building technique partial change among independent modeling.