Estimation of Whipple distribution parameters using the ordinary least squares method and Weighted least-squares method (comparative study)

Abstract

This research deals with estimating the shape parameter and the measurement
parameter for the two-parameter Weibull distribution function using a linear estimation
method to convert the distribution into a simple linear regression model and then using
the regular least squares method (OLS) and the weighted least squares method (WLS)
and compare them through the smallest mean squares error possible (MSE) In this
research, actual data were used to prepare the numbers of arrivals (different nationalities)
to Iraq through the Zarbatiya port, and to analyze and extract the results using the
MS.Excel2007 program. It was found that the weighted least squares method gave the
best estimator with the lowest average for the error box. The results are obtained.