We employ a probabilistic flood risk analysis to estimate exceedance probability curves, average annual loss (AAL) and probable maximum loss (PML) of households in the Danshuei River Basin in Taipei. We first establish a set of scenario precipitations, utilize the SOBEK model to simulate flood hazards and vulnerabilities, and estimate economic losses for each flood. The aggregate and occurrence exceedance probability curves for household losses are simulated using a Monte Carlo simulation. The average annual loss and the probable maximum loss in the Danshuei River Basin estimated using the aggregate exceedance probability curve can be applied to flood risk management and planning.