Least squares fitting the three - parameter inverse Weibull density ∗

The inverse Weibull model was developed by Erto [10]. In practice, the unknown parameters of the appropriate inverse Weibull density are not known and must be estimated from a random sample. Estimation of its parameters has been approached in the literature by various techniques, because a standard maximum likelihood estimate does not exist. To estimate the unknown parameters of the three-parameter inverse Weibull density we will use a combination of nonparametric and parametric methods. The idea consists of using two steps: in the first step we calculate an initial nonparametric density estimate which needs to be as good as possible, and in the second step we apply the nonlinear least squares method to estimate the unknown parameters. As a main result, a theorem on the existence of the least squares estimate is obtained, as well as its generalization in the lp norm (1 ≤ p < ∞). Some simulations are given to show that our approach is satisfactory if the initial density is of good enough quality. AMS subject classifications: 65D10, 62J02, 62G07, 62N05