4. Concluding remarks
In summary, the three most common techniques to estimate the parameters of a probability distribution, namely, the method of least squares (UWLS, WLS and QLS), the method of maximum likelihood (MLE) and the method of moments (ME) are compared to estimate the parameters of the shifted Gompertz distribution.
Numerical results via simulation studies conclude that the least squares method is a suitable and simple way to estimate the parameters of this probability distribution with an acceptable performance. The performance of the estimators obtained by the least squares method using weighting factors (WLS) is better than that of the UWLS estimators. The estimators obtained with more bias, especially for small sample sizes, are those of the method of moments.