Distribution fitting by using the mean of distances of empirical data

Yury Dodonov
2020 figshare.com  
A new method of distribution fitting based on the mean of distances from xi to other points of empirical data is herein presented. The distinctive feature of this new approach is that it functionally takes into account empirical data values without requiring "intuitive" quintile settings or complex mathematical calculations. An example for a normal distribution is presented. A plain simulation study shows that the new approach gives a more accurate estimate of the distribution parameters than
more » ... n parameters than the well-known and most used to date maximum likelihood estimation and maximum goodness-of-fit estimation methods, regardless of the data truncation strength. The proposed method is effective and easy to use and can be applied to any type of distribution, regardless of the shape and degree of the data truncation.
doi:10.6084/m9.figshare.13014410.v3 fatcat:6bs2k3btkfh6thszm2fvmzrgp4