Comparison of Resampling and Bayesian Approaches in Variance Component Estimation of a Hierarchical Univariate Mixed Effect Model

Doğan Narinç, Nihan Öksüz Narinç


The purpose of the study is to investigate the relative performance of two estimation procedures, a semi-frequentist estimation technique (via a Bootstrapped the restricted maximum likelihood: Bootstrap-REML) and Bayesian method (via a Gibbs sampler), for estimation of variance components of a two level hierarchical linear mixed model. For this purpose one variable named X was generated using R simulation with the structure of two level nested designs which showed Gaussian distribution. The variable X contains 10000 data, with an average of 0 and variances of 100 and. For this data, five different scenarios were created according to the rate of variance components and analyzes were carried out. All of the estimations and definitions of autocorrelation, changes of the total variance and estimation biases were performed for the posterior distributions and bootstrapped parameter distributions of all the scenarios. In general, the results obtained with both methods are close to each other, although the bias of the results obtained with the Gibbs sampling method was found less and autocorrelation was not found for Gibbs sampling estimates. In conclusion, according to the results of this study, it is not possible to say that using the Bootstrap-REML estimator under Gaussian distribution and balanced data is a good alternative to Bayesian Gibbs sampler.Perhaps different results may be obtained from another study using unbalanced data, non-normally distributed data and high sample sizes.


Bayesian; Gibbs Samplingt; VCE; REML; Bootstrap

Full Text:



Altay Y, Kılıç B, Aytekin İ, Keskin İ (2019). Determination of factors affecting mastitis in Holstein Friesian and Brown Swiss by using logistic regression analysis. Selcuk Journal of Agriculture and Food Sciences, 33(3), 194-197.

Aytekin İ, Doğan Ş, Odacı Ö, Gökcan G (2019). Estimation of variance components for birth and weaning weights in holstein-friesian calves by using WOMBAT software. Selcuk Journal of Agriculture and Food Sciences, 33(2), 88-93.

Browne WJ, Draper D (2006). A comparison of Bayesian and likelihood-based methods for fitting multilevel models. Bayesian analysis, 1(3), 473-514.

Dağ B, Keskin İ, Zülkadir U, Boztepe S (2003). Çumra Ziraat Meslek Lisesinde (Konya) yetiştirilen Esmer ineklerin süt verim özellikleri ve bu özelliklere ait tekrarlanma dereceleri. Selçuk Üniversitesi Ziraat Fakültesi Dergisi, 17(31), 13-17.

Delpish A (2009). Comparison of Restricted Maximum Like-lihood and Bootstrap via Minimum Norm Quadratic Un-biased Estimators for Hierarchical Linear Models under X ½ Assumptions. Journal of Statistical Research, 43(1), 69.

Efron B, Tibshirani R (1993) An Introduction to the Bootstrap. New York: Chapman and Hall.

Fırat MZ (1996a). Some Aspects of the Bayesian Approach to Statistical Modelling in Animal Breeding. Turkish Journal Of Veterinary And Animal Sciences, 20(5), 353-358.

Fırat MZ (1996b). Bayesian methods in animal breeding with different prior specification using gibbs sampling. Turkish Journal Of Veterinary And Animal Sciences, 20(5), 359-365.

Fırat MZ (2000). Comparative investigation of variance analysis, maximum l1kellhood andrestricted maximum l1kellhood for the estimation of variance components in balanced two-stage random nested designs. AnadoluÜn-iversitesiBilimveTeknolojiDergisi, 1, 105-113.

Gökmen M, Sarıyel V, Keskin İ, Dağ B (2008). Konya Meri-nosu koyunlarda bazı çevre faktörlerinin gebelik süresine etkileri. Selcuk Journal of Agriculture and Food Sciences, 22(44), 26-29.

Hadfield JD (2010). MCMC methods for multi-response generalized linear mixed models: the MCMC glmm R package. Journal of Statistical Software, 33(2), 1-22.

Hartley HO, Rao JN (1967). Maximum-likelihood estimation for the mixed analysis of variance model. Biometrika, 54(1-2), 93-108.

Maas, CJM, Hox, JJ (2005). Sufficient Sample sizes for Multilevel Modeling Methodology, 1(3), 85–91.

Narinç D, Karaman E, Fırat MZ, Aksoy T (2011). Estimation of multiple-trait genetic parameters and BLUP using dif-ferent estimation methods for some egg traits in Japanese quails. Kafkas Üniversitesi Veteriner Fakültesi Dergisi, 17(1), 117-123.

Patterson HD, Thompson R (1971). Recovery of inter-block information when block sizes are unequal. Biometrika, 58(3), 545-554.

Rasch D, Masata O (2006). Methods of variance component estimation. Czech Journal of Animal Science, 51(6), 227.

Raudenbush SW, Bryk AS (1986). A hierarchical model for studying school. Sociology of Education, 59(1), 1–17.

Robinson DL (1987). Estimation and use of variance compo-nents. Journal of the Royal Statistical Society: Series D (The Statistician), 36(1), 3-14.

SAS Institute. SAS version 9.4 In. Cary, NC: SAS Institute; (2013).

Searle SR, Casella G, McCulloch CE (2006). Variance Com-ponents. Hoboken. John Wiley & Sons, Inc. p. 1–501

Swallow WH, Monahan JF (1984). Monte Carlo comparison of ANOVA, MIVQUE, REML, and ML estimators of variance components. Technometrics, 26(1), 47-57.

Theobald CM, FIrat MZ, Thompson R (1997). Gibbs sam-pling, adaptive rejection sampling and robustness to prior specification for a mixed linear model [Bayesian inference, sire model]. Genetics Selection Evolution, 29(1), 57.

Wagenmakers EJ, Lee M, Lodewyckx T, Iverson GJ (2008). Bayesian versus frequentist inference. In Bayesian evalu-ation of informative hypotheses(pp. 181-207). Springer, New York, NY.

Zülkadir U, Şahin Ö, Aytekin İ, Boztepe S (2008). Malya kuzularda canlı ağırlık ve bazı vücut ölçülerinin tekrar-lanma dereceleri. Selçuk Tarım Bilimleri Dergisi, 22(45), 89-93.

Zülkadir U, Aytekin İ (2009). Genetic analysis of test day milk yields of Brown Swiss cattle raised at Konuklar State Farm in Turkey using MTDFREML. South African Journal of Animal Science. 39(1):10-14.



  • There are currently no refbacks.

Creative Commons Lisansı
Bu eser Creative Commons Alıntı-GayriTicari-Türetilemez 4.0 Uluslararası Lisansı ile lisanslanmıştır.