Topic: Contributed on Model asessment Title: Estimating the Zolotarev distance to the class of exponential distributions Authors:  Amparo Baillo - Universidad Autonoma de Madrid (Spain) [presenting]
Javier Carcamo - Universidad Autonoma de Madrid (Spain)
Abstract: The exponential distribution is one of the most important and widely used distributions in statistical applications. Knowing the distance from a positive random variable $X$ to the exponential class $\mathcal{E}$ would assess if it is reasonable or not to assume an exponential model as an approximation to the unknown probability distribution of $X$. We study the asymptotic behaviour of a plug-in estimator of a normalized version of the Zolotarev distance between $X$ and $\mathcal{E}$. Its performance in practice has been checked via simulations and the analysis of real data sets.