Obtaining daily precipitation parameters from meteorological yearbooks

This study compares the performance of two different methods, one based on the method of moments and the other based on maximum likelihood, in their assessment of the main statistical features of daily precipitation using censored information found in meteorological yearbooks. A Monte Carlo simulation was performed to generate series of daily rainfall and summarize them in the same format meteorological yearbooks report precipitation. While the method of moments for censored data (c-Mo) is a good alternative for the estimation of the parameters of precipitation occurrence due to its simplicity, the method of maximum likelihood for censored data (c-MLE) shows better results in estimating the parameters and quantiles of daily precipitation intensities. The c-MLE method is a good alternative even in situations in which the wrong candidate distribution is fit. The precision of the estimates obtained using both methods is increased when larger sample sizes are used, however c-MLE shows better results than c-Mo method for small sample sizes.