Metody nieparametryczne jako podstawa estymacji modeli ekonometrycznych z odstającymi obserwacjami

In this paper the authors present a nonparametric method of estimating the parameters of the linear econometric model, which is the method of least absolute deviations (LAD). The aim of this article is to examine the extent to which the parameter estimation method of least absolute deviations is resistant to changes in parameter values in case of outliers. In this paper, a hypothetical example examines the impact of the so-called outliers on the model parameters estimated by OLS methods and LAD respectively. In addition, the work raises the problem of the use of permutation methods MRPP in testing certain statistical hypotheses when the distributions of examined random variables distributions are not normal.