A maximum likelihood estimation approach for spliced distributions obtained through quantile splicing

Abstract

This mini-dissertation proposes constructing a family of spliced distributions at a point different from the median, hence k=1/4 instead of k=1/2, using the method of quantile splicing proposed by Mac'Oduol et al. (2020). General results of these families of distributions are developed and the maximum likelihood approach is explored and investigated for estimation purposes. Moreover, a numerical application is presented in order to illustrate the implementation and application of the proposed method.