Statistical Risk Assessment Framework for Bank Credit Portfolio Management in Nigerian Commercial Banks

Credit portfolio risk management in Nigerian commercial banks requires rigorous statistical frameworks that account for loan concentration, default correlation, and macroeconomic sensitivity to support regulatory capital adequacy and provisioning decisions. This study develops and applies a statistical credit risk assessment framework to the loan portfolios of five tier-1 Nigerian commercial banks using published financial statements and CBN credit bureau data. Value at Risk for credit portfolios was estimated using the CreditMetrics methodology with Monte Carlo simulation of 100,000 scenarios. Default probabilities were derived from historical non-performing loan transition matrices. Asset value correlations were estimated from equity return data. Expected credit loss and unexpected credit loss were computed by sector and customer segment. The mean one-year portfolio Expected Credit Loss across the five banks was 3.4 percent of gross loans. Monte Carlo simulation identified oil and gas sector concentration as the primary driver of tail risk, with 99 percent Value at Risk sensitivity increasing by 2.8 times under oil price stress scenarios. Retail loan portfolios showed lower correlation-adjusted unexpected credit loss than corporate segments. Stress testing under exchange rate devaluation scenarios increased portfolio Expected Credit Loss by 1.9 percentage points. The study provides a professionally deployable credit risk statistical framework and recommends sector concentration limits and stress-tested capital buffer requirements for Nigerian commercial banks. Keywords: credit risk, Value at Risk, Monte Carlo simulation, Nigerian banks, portfolio management