The article explores how to estimate the risk of large financial losses in credit portfolios. The researchers used Monte Carlo simulation to create a probability distribution for portfolio losses, then applied Extreme Value Theory to analyze the tail of this distribution. By using bootstrap techniques, they calculated the Value-at-Risk (VaR) and found that the EVT-based approach resulted in narrower confidence intervals and less sampling uncertainty compared to the empirical method. Additionally, the bootstrap replicate's distribution for the EVT-based method showed a better shape than the empirical method.