Comparsion between weighted quadratic loss function and quadratic loss function to estimate asymmetric Laplace distribution parameters

Radi, Asmaa Khamis; Abdullah, Ebtisam Karim

doi:10.22075/ijnaa.2022.6030

Comparsion between weighted quadratic loss function and quadratic loss function to estimate asymmetric Laplace distribution parameters

Document Type : Research Paper

Authors

Asmaa Khamis Radi

Ebtisam Karim Abdullah

University of Baghdad, College of Administration and Economics, Department of Statistics, Iraq.

10.22075/ijnaa.2022.6030

Abstract

The skewness and scale parameters of the asymmetric Laplace distribution are estimated with Bayesian methods using quadratic loss function and the weighted quadratic loss function, respectively, based on the functions of the prior of the gamma distribution and the exponential distribution for each of the skewness and scale parameters. These estimates were compared using integral mean square error, which was based on the real data technique of the stock prices the Iraqi market. The results revealed that the bayes estimator outperformed the quadratic loss function under the weighted quadratic loss function.

Keywords

Asymmetric Laplace distribution

Bays estimator underweighted quadratic loss function

Quadratic loss function

Lindley approximation

[1] J. K. Ghosh, M. Delampady and T. Samanta, An Introduction to Bayesian Analysis: Theory and Methods,
Springer , first Edition, 2006.
[2] M. R. Hasan and A. R. Baizid,Bayesian Estimation under Different Loss Function Using Gamma prior for the
Case of Exponential Distribution ,J. Sci. Res. , 9 (1)(2017) 67-78 .
[3] S. Kotz, T. Kozubowski and K. Podgorski, The Laplace Distribution and Generalizations: A revisit with Applications to Communications, Economics, Engineering, and Finance, Springer, Science and Business Media,
2001.
[4] T. J. Kozubowski and K. Podg´orski ,Asymmetric Laplace Laws and Modeling Financial Data, Math. Comput.
Modell., 34 (2001) 1003-1021 .
[5] H. Rahman, M. K. Roy and A. R. Baizid, Bayes Estimation Under Conjugate Prior for the Case of Power
Function Distribution, Am. J. Math. Stat., 2 (3) (2012) 44-48 .
[6] H. A. Rasheed and Z. N. Khalifa, Some Bayes Estimators for Maxwell Distribution by Using New Loss Function,
Al-Mustansiriyah J. Sci., 28 (1)( 2017) 103-111.
[7] H. Sultan and S. P. Ahmad , Bayesian Analysis for Generalized Rayleigh Distribution: Lindley’s approximation,Int. J. Adv. Res. Sci. Eng.,6(3)(2017) 214-223 .