In this study, regional frequency analysis of extreme rainfalls in Hong Kong is presented using the index-event L-moment based method of Hosking and Wallis. Data analyzed are annual maximum rainfall series of different durations (from 5-min to 1440-min) at a total of 122 automatic raingauges with record lengths varying from 10 to 26 years. The entire territory of Hong Kong was divided into three to five hydrologically homogeneous sub-regions according to geographical attributes (latitude, longitude, and elevation) and mean annual rainfall. Goodness-of-fit Z-test and L-moment ratio diagram were used to select the best-fit regional distributions from five three-parameter probability distributions (GEV, GLO, GPA, LN3, and PE3) and the five-parameter Wakeby distribution. Several factors associated with the sample data used in regional frequency analysis (varying site record lengths, number of sites within region, and regional heterogeneity) were also studied through Monte Carlo simulation.