Weighted Likelihood Estimator for Exponentiated Weibull Distribution

Main Article Content

Kanlaya Boonlha

Abstract

Abstract


In this study, we apply the weighted likelihood estimator (WLE) to the Exponentiated Weibull EW distribution with contamination. And we also compare the performance to the maximum likelihood estimator (MLE). We fix the central distribution to be the EW distribution gif.latex?(\alpha,\beta,\theta)with(2,1,1) and (2,2,2) and the contamination is EW distribution with parameter gif.latex?\alpha_{1}=\alpha(1+\Delta),&space;\beta_{1}=\beta(1+\Delta),\alpha_{1}=\theta_{1}(\theta+\Delta), where gif.latex?\Delta= 1, 5 and the contamination proportion ( gif.latex?\epsilon )= 0.01, 0.03, and 0.05 and the values of pre–assigned small probability k= 0.01, 0.03, 0.05 with shape parameter gif.latex?\alpha= 2. We perform a Monte Carlo simulation to compare the perform of MLE and WLE. The simulation results are based on the 10,000 replace. The efficiency of the MLE and WLE are compared based on the bias values and the root mean square error (RMSE).  The result show that the sample size increase as the bias and RMSE of the MLE and WLE decrease in all most of the cases. The WLE method for gif.latex?\theta  provide better than MLE result in the term of the bias and RMSE when k is large for the estimator scale parameter (gif.latex?\theta ).  While the MLE method for    gif.latex?\betaprovide better WLE result in the term of the bias and RMSE for the estimator shape parameter (gif.latex?\beta).


 

Article Details

How to Cite
Boonlha, K. (2024). Weighted Likelihood Estimator for Exponentiated Weibull Distribution. Interdisciplinary Research Review, 19(5). Retrieved from https://ph02.tci-thaijo.org/index.php/jtir/article/view/247829
Section
Research Articles

References

Mudholkar, Govind S., Deo Kumar Srivastava, and Marshall Freimer. "The exponentiated Weibull family: A reanalysis of the bus-motor-failure data." Technometrics 37.4 (1995): 436-445.

Gupta, Rameshwar D., and Debasis Kundu. "Exponentiated exponential family: an alternative to gamma and Weibull distributions." Biometrical Journal: Journal of Mathematical Methods in Biosciences 43.1 (2001): 117-130.

Nassar, Manal M., and Fathy H. Eissa. "On the exponentiated Weibull distribution." Communications in Statistics-Theory and Methods 32.7 (2003): 1317-1336.

Sobhi, Mashail M. AL, and Ahmed A. Soliman. "Estimation for the exponentiated Weibull model with adaptive Type-II progressive censored schemes." Applied Mathematical Modelling 40.2 (2016): 1180-1192.

Jing, Zhao, et al. "Performance analysis for mixed FSO/RF Nakagami-m and Exponentiated Weibull dual-hop airborne systems." Optics Communications 392 (2017): 294-299.

Khan, R. U., Zaki Anwar, and Haseeb Athar. "Recurrence relations for single and product moments of dual generalized order statistics from exponentiated Weibull distribution." Aligarh J. Statist 28 (2008): 37-45.

Pal, Manisha, M. Masoom Ali, and Jungsoo Woo. "Exponentiated weibull distribution." Statistica 66.2 (2006): 139-147.

Boudt, Kris, Derya Caliskan, and Christophe Croux. "Robust explicit estimators of Weibull parameters." Metrika 73.2 (2011): 187-209.

Ahmed, Ejaz S., Andrei I. Volodin, and Abdulkadir A. Hussein. "Robust weighted likelihood estimation of exponential parameters." IEEE Transactions on reliability 54.3 (2005): 389-395.

Hu, Feifang, and James V. Zidek. "The relevance weighted likelihood with applications." Empirical Bayes and Likelihood Inference. Springer, New York, NY, 2001. 211-235.

Nichols, Michele D., and W. J. Padgett. "A bootstrap control chart for Weibull percentiles." Quality and reliability engineering international 22.2 (2006): 141-151.