AN EMPIRICAL STUDY OF ECONOMIC VARIABLES AND CRIME FIGURES IN NIGERIA USING MULTIVARIATE TECHNIQUES

This study applies multivariate analysis techniques to investigate key Gross Domestic Product (GDP) and crime variables in Nigeria, aiming to identify those with positive impacts on society. Using yearly crime data from 2012 to 2020 and quarterly GDP data spanning 1981Q1 to 2019Q1, the analysis employed Principal Components Analysis (PCA), Factor Analysis (FA), and Cluster Analysis (CA) via R statistical software. Three rotation methods within PCA were applied to capture variability in both datasets, followed by cluster identification using four approaches to group variables into more homogeneous clusters. Significant differences were observed between None and Varimax rotation methods compared to Promax rotation for economic variables, whereas crime data showed no significant differences across rotation methods regardless of sample size. Factor analysis revealed that principal components explained between 78.2% and 99.5% of the variability in economic variables, and between 17.7% and 99.5% in large-sample crime data, and 18.1% to 99.3% in small-sample crime data. The study identified seven clusters for economic variables, seven for large-sample crime data, and five for small-sample crime data, providing insights into the structure and relationships among these variables