Published

2016-01-01

Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values

Estimación shrinkage de los parámetros de la distribución exponencial basada en valores record

DOI:

https://doi.org/10.15446/rce.v39n1.55137

Keywords:

Exponential Distribution, Minimax Regret, Record Value, Risk Function, Shrinkage Estimator (en)
Estimador shrinkage, Distribución exponencial, Minimax regret, Función de riesgo, Valor record. (es)

Downloads

Authors

  • Hojatollah Zakerzadeh Yazd University, Yazd, Iran
  • Ali Akbar Jafari Yazd University, Yazd, Iran
  • Mahdieh Karimi Yazd University, Yazd, Iran

This paper studies shrinkage estimation after the preliminary test for the parameters of exponential distribution based on record values. The optimal value of shrinkage coefficients is also obtained based on the minimax regret criterion. The maximum likelihood, pre-test, and shrinkage estimators are compared using a simulation study. The results to estimate the scale parameter show that the optimal shrinkage estimator is better than the maximum likelihood estimator in all cases, and when the prior guess is near the true value, the pre-test estimator is better than shrinkage estimator. The results to estimate the location parameter show that the optimal shrinkage estimator is better than maximum likelihood estimator when a prior guess is close to the true value. All estimators are illustrated by a numerical example.

Este artículo estudia la estimación shrinkage posterior al test preliminar de los parámetros de la distribución exponencial basada en valores record. El valor óptimo de los coeficientes de shrinkage es obtenido también usando el criterio minimax regret. La máxima verosimilitud, pre-test, y los estimadores shrinkage son obtenidos usando estudios de simulación. Los resultados de la estimación del parámetro de escala muestran que el estimador shrinkage es major que el de máxima verosimilitud en todos los casos, y cuando el valor a priori es cercano del valor real, el estimador pre-test es mejor que el estimador shrinkage. Los resultados de estimación del parámetro de localización muestran que el estimador de shrinkage óptimo es major que el de máxima verosimilitud cuando el valor a priori es cercano al real. Todos los estimadores son ilustrados con un ejemplo numérico.

Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values

Estimación shrinkage de los parámetros de la distribución exponencial basada en valores record

HOJATOLLAH ZAKERZADEH1, ALI AKBAR JAFARI2, MAHDIEH KARIMI3

1Yazd University, Department of Statistics, Yazd, Iran. Assistant Professor. Email: hzaker@yazd.ac.ir
2Yazd University, Department of Statistics, Yazd, Iran. Assistant Professor. Email: aajafari@yazd.ac.ir
3Yazd University, Department of Statistics, Yazd, Iran. Student. Email: mkarimiz68@yahoo.com


Abstract

This paper studies shrinkage estimation after the preliminary test for the parameters of exponential distribution based on record values. The optimal value of shrinkage coefficients is also obtained based on the minimax regret criterion. The maximum likelihood, pre-test, and shrinkage estimators are compared using a simulation study. The results to estimate the scale parameter show that the optimal shrinkage estimator is better than the maximum likelihood estimator in all cases, and when the prior guess is near the true value, the pre-test estimator is better than shrinkage estimator. The results to estimate the location parameter show that the optimal shrinkage estimator is better than maximum likelihood estimator when a prior guess is close to the true value. All estimators are illustrated by a numerical example.

Key words: Exponential Distribution, Minimax Regret, Record Value, Risk Function, Shrinkage Estimator.


Resumen

Este artículo estudia la estimación shrinkage posterior al test preliminar de los parámetros de la distribución exponencial basada en valores record. El valor óptimo de los coeficientes de shrinkage es obtenido también usando el criterio minimax regret. La máxima verosimilitud, pre-test, y los estimadores shrinkage son obtenidos usando estudios de simulación. Los resultados de la estimación del parámetro de escala muestran que el estimador shrinkage es major que el de máxima verosimilitud en todos los casos, y cuando el valor a priori es cercano del valor real, el estimador pre-test es major que el estimador shrinkage. Los resultados de estimación del parámetro de localización muestran que el estimador de shrinkage óptimo es major que el de máxima verosimilitud cuando el valor a priori es cercano al real. Todos los estimadores son ilustrados con un ejemplo numérico.

Palabras clave: estimador shrinkage, distribución exponencial, minimax regret, función de riesgo, valor record.


Texto completo disponible en PDF


References

1. Ahsanullah, M. (1995), Record statistics, Nova Science Publishers Commack, New York.

2. Arnold, B., Balakrishnan, N. & Nagaraja, H. (1998), Records, John Wiley & Sons Inc, New York.

3. Balakrishnan, N. & Chan, P. S. (1994), 'Record values from Rayleigh and Weibull distributions and associated inference', National Institute of Standards and Technology Journal of Research, Special Publication, Proceedings of the Conference on Extreme Value Theory and Applications 866, 41-51.

4. Bancroft, T. A. (1944), 'On biases in estimation due to the use of preliminary tests of significance', The Annals of Mathematical Statistics 15(2), 190-204.

5. Bancroft, T. A. & Han, C. P. (1977), 'Inference based on conditional specification: a note and a bibliography', International Statistical Review 45, 117-127.

6. Bhattacharya, S. K. & Srivastava, V. K. (1974), 'A preliminary test procedure in life testing', Journal of the American Statistical Association 69(347), 726-729.

7. Chiou, P. (1990), 'Estimation of scale parameters for two exponential distributions', IEEE Transactions on Reliability 39(1), 106-109.

8. Chiou, P. & Han, C. P. (1995), 'Conditional interval estimation of the exponential location parameter following rejection of a pre-test', Communications in Statistics-Theory and Methods 24(6), 1481-1492.

9. Chiou, P. & Han, C. (1989), 'Shrinkage estimation of threshold parameter of the exponential distribution', IEEE transactions on reliability 38, 449-453.

10. Chiou, P. & Miao, W. (2005), 'Shrinkage estimation for the difference between exponential guarantee time parameters', Computational Statistics and Data Analysis 48(3), 489-507.

11. Dunsmore, I. R. (1983), 'The future occurrence of records', Annals of the Institute of Statistical Mathematics 35(1), 267-277.

12. Han, Chien-Pai, Rao, C. & Ravichandran, J. (1988), 'Inference based on conditional speclfication', Communications in Statistics-Theory and Methods 17(6), 1945-1964.

13. Inada, K. (1984), 'A minimax regret estimator of a normal mean after preliminary test', Annals of the Institute of Statistical Mathematics 36(1), 207-215.

14. Pandey, B. N. (1983), 'Shrinkage estimation of the exponential scale parameter', IEEE Transactions on Reliability 32(2), 203-205.

15. Pandey, B. N. (1997), 'Testimator of the scale parameter of the exponential distribution using LINEX loss function', Communications in Statistics-Theory and Methods 26(9), 2191-2202.

16. Prakash, G. & Singh, D. (2008), 'Shrinkage estimation in exponential type-II censored data under LINEX loss', Journal of the Korean Statistical Society 37(1), 53-61.

17. Zakerzadeh, H. & Karimi, M. (2013), Preliminary test estimation of exponential distribution based on record values, 'The 9th Seminar on Probability and Stochastic Processes, Iran, Zahedan'.


[Recibido en agosto de 2014. Aceptado en febrero de 2015]

Este artículo se puede citar en LaTeX utilizando la siguiente referencia bibliográfica de BibTeX:

@ARTICLE{RCEv39n1a03,
    AUTHOR  = {Zakerzadeh, Hojatollah and Jafari, Ali Akbar and Karimi, Mahdieh},
    TITLE   = {{Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values}},
    JOURNAL = {Revista Colombiana de Estadística},
    YEAR    = {2016},
    volume  = {39},
    number  = {1},
    pages   = {33-44}
}

References

Ahsanullah, M. (1995), Record statistics, Nova Science Publishers Commack, New York.

Arnold, B., Balakrishnan, N. & Nagaraja, H. (1998), Records, John Wiley & Sons Inc, New York.

Balakrishnan, N. & Chan, P. S. (1994), ‘Record values from Rayleigh and Weibull distributions and associated inference’, National Institute of Standards and Technology Journal of Research, Special Publication, Proceedings of the Conference on Extreme Value Theory and Applications 866, 41–51.

Bancroft, T. A. (1944), ‘On biases in estimation due to the use of preliminary tests of significance’, The Annals Bancroft, T. A. & Han, C. P. (1977), ‘Inference based on conditional specification: a note and a bibliography’, International Statistical Review 45, 117–127.

Bancroft, T. A. & Han, C. P. (1977), ‘Inference based on conditional specification: a note and a bibliography’, International Statistical Review 45, 117–127.

Bhattacharya, S. K. & Srivastava, V. K. (1974), ‘A preliminary test procedure in life testing’, Journal of the American Statistical Association 69(347), 726–729.

Chiou, P. (1990), ‘Estimation of scale parameters for two exponential distributions’, IEEE Transactions on Reliability 39(1), 106–109.

Chiou, P. & Han, C. (1989), ‘Shrinkage estimation of threshold parameter of the exponential distribution’, IEEE transactions on reliability 38, 449–453.

Chiou, P. & Han, C. P. (1995), ‘Conditional interval estimation of the exponential location parameter following rejection of a pre-test’, Communications in Statistics-Theory and Methods 24(6), 1481–1492.

Chiou, P. & Miao, W. (2005), ‘Shrinkage estimation for the difference between exponential guarantee time parameters’, Computational Statistics and Data Analysis 48(3), 489–507.

Dunsmore, I. R. (1983), ‘The future occurrence of records’, Annals of the Institute of Statistical Mathematics 35(1), 267–277.

Han, C.-P., Rao, C. & Ravichandran, J. (1988), ‘Inference based on conditional speclfication’, Communications in Statistics-Theory and Methods 17(6), 1945–1964.

Inada, K. (1984), ‘A minimax regret estimator of a normal mean after preliminary test’, Annals of the Institute of Statistical Mathematics 36(1), 207–215.

Pandey, B. N. (1983), ‘Shrinkage estimation of the exponential scale parameter’, IEEE Transactions on Reliability 32(2), 203–205.

Pandey, B. N. (1997), ‘Testimator of the scale parameter of the exponential distribution using LINEX loss function’, Communications in Statistics-Theory and Methods 26(9), 2191–2202.

Prakash, G. & Singh, D. (2008), ‘Shrinkage estimation in exponential type-II censored data under LINEX loss’, Journal of the Korean Statistical Society 37(1), 53–61.

Zakerzadeh, H. & Karimi, M. (2013), Preliminary test estimation of exponential distribution based on record values, in ‘The 9th Seminar on Probability and Stochastic Processes, Iran, Zahedan’. of Mathematical Statistics 15(2), 190–204.

How to Cite

APA

Zakerzadeh, H., Jafari, A. A. and Karimi, M. (2016). Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values. Revista Colombiana de Estadística, 39(1), 33–44. https://doi.org/10.15446/rce.v39n1.55137

ACM

[1]
Zakerzadeh, H., Jafari, A.A. and Karimi, M. 2016. Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values. Revista Colombiana de Estadística. 39, 1 (Jan. 2016), 33–44. DOI:https://doi.org/10.15446/rce.v39n1.55137.

ACS

(1)
Zakerzadeh, H.; Jafari, A. A.; Karimi, M. Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values. Rev. colomb. estad. 2016, 39, 33-44.

ABNT

ZAKERZADEH, H.; JAFARI, A. A.; KARIMI, M. Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values. Revista Colombiana de Estadística, [S. l.], v. 39, n. 1, p. 33–44, 2016. DOI: 10.15446/rce.v39n1.55137. Disponível em: https://revistas.unal.edu.co/index.php/estad/article/view/55137. Acesso em: 19 apr. 2024.

Chicago

Zakerzadeh, Hojatollah, Ali Akbar Jafari, and Mahdieh Karimi. 2016. “Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values”. Revista Colombiana De Estadística 39 (1):33-44. https://doi.org/10.15446/rce.v39n1.55137.

Harvard

Zakerzadeh, H., Jafari, A. A. and Karimi, M. (2016) “Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values”, Revista Colombiana de Estadística, 39(1), pp. 33–44. doi: 10.15446/rce.v39n1.55137.

IEEE

[1]
H. Zakerzadeh, A. A. Jafari, and M. Karimi, “Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values”, Rev. colomb. estad., vol. 39, no. 1, pp. 33–44, Jan. 2016.

MLA

Zakerzadeh, H., A. A. Jafari, and M. Karimi. “Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values”. Revista Colombiana de Estadística, vol. 39, no. 1, Jan. 2016, pp. 33-44, doi:10.15446/rce.v39n1.55137.

Turabian

Zakerzadeh, Hojatollah, Ali Akbar Jafari, and Mahdieh Karimi. “Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values”. Revista Colombiana de Estadística 39, no. 1 (January 1, 2016): 33–44. Accessed April 19, 2024. https://revistas.unal.edu.co/index.php/estad/article/view/55137.

Vancouver

1.
Zakerzadeh H, Jafari AA, Karimi M. Optimal Shrinkage Estimations for the Parameters of Exponential Distribution Based on Record Values. Rev. colomb. estad. [Internet]. 2016 Jan. 1 [cited 2024 Apr. 19];39(1):33-44. Available from: https://revistas.unal.edu.co/index.php/estad/article/view/55137

Download Citation

CrossRef Cited-by

CrossRef citations4

1. Ayman Baklizi. (2017). Preliminary Test Estimation of the Threshold in the Two-Parameter Exponential Distribution Based on Records and Minimax Regret Significance Levels. American Journal of Mathematical and Management Sciences, 36(3), p.196. https://doi.org/10.1080/01966324.2017.1310005.

2. N. J. Hassan, J. Mahdi Hadad, A. Hawad Nasar. (2020). Bayesian Shrinkage Estimator of Burr XII Distribution. International Journal of Mathematics and Mathematical Sciences, 2020, p.1. https://doi.org/10.1155/2020/7953098.

3. Housila P. Singh, Harshada Joshi, Gajendra K. Vishwakarma. (2024). Shrinkage estimation of θα in gamma density G(1/θ, p) using prior information. Journal of Engineering Mathematics, 144(1) https://doi.org/10.1007/s10665-023-10329-9.

4. Housila P. Singh, Gajendra K. Vishwakarma, Harshada Joshi, Shubham Gupta. (2024). Shrinkage estimation for square of location parameter of the exponential distribution with known coefficient of variation. Journal of Computational and Applied Mathematics, 437, p.115489. https://doi.org/10.1016/j.cam.2023.115489.

Dimensions

PlumX

Article abstract page views

276

Downloads

Download data is not yet available.