ABSTRACT
This chapter discusses some of the common methods of parameter estimation. These include the method of moments, the maximum likelihood method, the probability weighted moments method, the least squares method, maximum entropy, mixed moments, the generalized method of moments, and incomplete means method. Three of the more commonly used methods are the method of moments, the maximum likelihood method and the probability weighted moments method. The estimated parameters are used to calculate quantile estimates for different return periods or, conversely, to calculate the return period for a given flood magnitude. There are two types of error associated with quantile estimates. The first type arises from the assumption that the observed data follow a particular distribution. The second type is the error inherent in parameters estimated from small samples. This error can be reduced by using a method which gives minimum variance parameter estimates, which in turn would result in the smallest variance in quantile estimates.
