The MLE- Distributions Fitting for Detecting Extreme Points, and Possible Repetitions of Sudden Breaking Points in Data. Conference Proceedings Federal University of Technology, Minna 3 RD School of Physical Sciences Biennial International Conference

Abstract

This work reports on how to use the probability graphical method via the maximum likelihood estimation (MLE) approach to obtain the various pivoting model’s parameters that gives the higher estimate value among the class of unbiased estimates. The probability distribution fitting was implemented via the graphical method. This helps to detect various outliers, intervals and the possible repetitions of sudden breaking points. The fitting was also obtained to predict and help in assessments during breakdowns in economic activities. Self-written R code and the Easy fit software were used to fit the household income data to suggest the possible probability distribution(s) for the data. The distributions were taken as the functional form of the income’s (X as an r.v) probability distribution and they were empirically solved using the MLE method. The estimate that is most consistent with the sample data was solved computationally based on the distribution function(s).