Big Data Analytics and Predictive Underwriting in the Nigerian Insurance Market: Opportunities and Governance Challenges

This dissertation investigates the adoption of big data analytics and predictive modeling in insurance underwriting in Nigeria, analyzing both the value creation potential and the governance challenges associated with algorithmic underwriting in the Nigerian regulatory and social context. Advanced analytics including machine learning-based risk scoring, telematics data utilization, social media behavioral signals, and geospatial risk mapping are transforming underwriting precision in global insurance markets, but their application in the Nigerian context raises unique questions about data quality, algorithmic fairness, privacy rights, and regulatory capacity. The study employs a multidisciplinary design integrating insurance economics, data science, and technology governance frameworks. A primary quantitative experiment develops and tests a machine learning-based underwriting model for motor insurance using claims and telematics data from a partnering insurance company, comparing its predictive performance against traditional rating factor models. A parallel governance analysis employs policy analysis and stakeholder consultation methods to map the regulatory gaps in Nigeria's data protection and insurance algorithmic governance framework. In-depth interviews with 35 data scientists, underwriting managers, and regulators assess implementation readiness and governance concerns. The study develops an original Algorithmic Underwriting Governance Framework (AUGF) for Nigeria as its primary policy contribution. Keywords: Big Data Analytics, Predictive Underwriting, Machine Learning, Insurance Governance, Nigeria.