Abstract
In this study, parameters of the one-way analysis of variance (ANOVA) model based on the type II censored samples are estimated when the distribution of the error terms is Jones and Faddy’s skew t. In the estimation procedure, maximum likelihood (ML) methodology is used. Solutions of the likelihood equations are obtained using an iterative method because of the nonlinear terms existing in the likelihood equations. A modified version of the ML methodology is also used. It gives the explicit estimators of the model parameters and has no computational difficulties. A new test statistic is proposed for testing the equality of the treatment effects. An extensive Monte Carlo simulation study is carried out to compare the powers of the test statistics based on the ML and the modified ML estimators of the parameters. Real data are analyzed at the end of the study to illustrate the implementation of the proposed methodologies.
Similar content being viewed by others
References
Acitas S, Kasap P, Senoglu B, Arslan O (2013a) One-step M-estimators: Jones and Faddy’s skew t distribution. J Appl Stati 40(7):1545–1560
Acitas S, Kasap P, Senoglu B, Arslan O (2013b) Robust estimation with the skew \(t_{2}\) distribution. Pak J Stat 29(4):409–430
Acitas S, Senoglu B, Arslan O (2015) Alpha-skew generalized t distribution. Revista Colombiana de Estadística 38(2):353–370
Álvarez BL, Gamero MDJ (2012) A note on bias reduction of maximum likelihood estimates for the scalar skew t distribution. J Stat Plan Inference 142:608–612
Arellano-Valle RB, Genton MG (2005) On fundamental skew distributions. J Multivar Anal 96:93–116
Arslan O (2009) Maximum lkelihood parameter estimation for the multivariate skew-slash distribution. Stat Probab Lett 79:2158–2165
Arslan T (2015) Robust estimation of the model parameters in experimental design for the complete and censored samples. Ph.D. dissertation. Eskisehir Osmangazi University, Eskisehir, Turkey
Arslan T, Senoglu B (2016) Statistical inference for the location model when the distribution of the error terms is Jones and Faddys skew t: type II censored samples. In: International conference on trends and perspectives in linear statistical inference (LINSTAT16), 22–25 Aug 2016, Istanbul, Turkey, p 28
Arslan T, Senoglu B (2017) Estimation for the location and the scale parameters of the Jones and Faddy’s skew t distribution under the doubly type II censored samples. Anadolu Univ J Sci Technol B Theor Sci 5(1):100–110
Barnett VD (1966) Evaluation of the maximum likelihood estimator when the likelihood equation has multiple roots. Biometrika 53:151–165
Bartlett MS (1953) Approximate confidence intervals. Biometrika 40:12–19
Basso RM, Lachos VH, Cabral CRB, Ghosh P (2010) Robust mixture modeling on scale mixtures of skew-normal distributions. Comput Stat Data Anal 54:2926–2941
Behboodian J, Jamalizadeh A, Balakrishnan N (2006) A new class of skew-Cauchy distributions. Stat Probab Lett 76:1488–1493
Box GEP, Cox DR (1964) An analysis of transformations (with disscussion). J R Stat Soc B26:211–252
Demirhan H, Kalaylioglu Z (2015) On the generalized multivariate Gumbel distribution. Stat Probab Lett 103:93–99
Flecher C, Naveau P, Allard D (2009) Estimating the closed skew-normal distribution parameters using weighted moments. Stat Probab Lett 79:1977–1984
García VJ, Gómez-Déniz E, Vázquez-Polo FJ (2010) A new skew generalization of the normal distribution: properties and applications. Comput Stat Data Anal 54:2021–2034
Gayen AK (1950) The distribution of the variance ratio in random samples of any size drawn from nonnormal universes. Biometrika 37:236–255
Geary RC (1932) Testing for normality. Biometrika 23:114–133
Harandi SS, Alamatsaz MK (2013) Alpha-skew Laplace distribution. Stat Probab Lett 83:774–782
Huang WJ, Chen YH (2007) Generalized skew-Cauchy distribution. Stat Probab Lett 77:1137–1147
Huber PJ (1981) Robust statistics. Wiley, New York
Islam MQ, Tiku ML (2004) Multiple linear regression model under nonnormality. Commun Stat Theory Methods 33(10):2443–2467
Jones MC (2008) The t family and their close and distant relations. J Korean Stat Soc 37:293–302
Jones MC, Faddy MJ (2003) A skew extension of the t-distribution, with applications. J R Stat Soc Ser B 65:159–174
Kendall MG, Stuart A (1961) The advanced theory of statistics, vol 2. Charles Griffin and Co., London
Kheradmandi A, Rasekh A (2015) Estimation in skew-normal linear mixed measurement error models. J Multivar Anal 136:1–11
Lin TI (2009) Maximum likelihood estimation for multivariate skew normal mixture models. J Multivar Anal 100:257–265
Mameli V (2015) The Kumaraswamy skew-normal distribution. Stat Probab Lett 104:75–81
Mudholkar GS, Hutson AD (2009) The epsilon-skew-normal distribution for analyzing near-normal data. J Stat Plan Inference 83:291–309
Nadarajah S, Ali MM (2004) A skew truncated T distribution. Math Comput Model 40:935–939
Pearson ES (1932) The analysis of variance in case of nonnormal variation. Biometrika 23:114–133
Puthenpura S, Sinha NK (1986) Modified maximum likkelihood method for the robust estimation of system parameters from very noise data. Automatica 22:231–235
Senoglu B (2005) Robust \(2^{k}\) factorial design with Weibull error distributions. J Appl Stat 32(10):1051–1066
Senoglu B (2007) Estimating parameters in one-way analysis of covarinace model with short-tailed symmetric error distributions. J Comput Appl Math 201:275–283
Senoglu B, Tiku ML (2001) Analysis of variance in exprimental design with nonnormal error distributions. Commun Stat Theory Methods 30(7):1335–1352
Senoglu B, Tiku ML (2004) Censored and truncated samples in experimental design under non-normality. Stat Methods 6(2):173–199
Senoglu B, Acitas S, Kasap P (2012) Robust \(2^{k}\) factorial designs: non-normal symmetric distribution. Pak J Stat 28(1):93–114
Srivastava ABL (1959) Effect of nonnormality on the power of the analysis of variance test. Biometrika 46:114–122
Tiku ML (1964) Approximating the general nonnormal variance ratio sampling distributions. Biometrika 51:83–95
Tiku ML (1967) Estimating the mean and standart deviation from censored normal sample. Biometrika 54:155–165
Tiku ML (1968) Estimating the parameters of normal and logistic distributions from censored normal sample. Aust J Stat 10:64–67
Tiku ML (1971) Power function of the F test under nonnormal situations. J Am Stat Assoc 66:913–916
Tiku ML (1973) Testing group effects from type II censored normal samples in experimental design. Biometrics 29:24–33
Tiku ML (1982) Testing linear contrast of means in experimental design without assuming normality and homogeneity of variances. Biometr J 24(6):613–627
Tiku ML, Stewart DE (1977) Estimating and testing group effects from type I censored normal samples in experimental design. Commun Stat Theory Methods A 6(15):1485–1501
Vaughan DC (1992) On the Tiku–Suresh method of estimation. Commun Stat Theory Methods 21(2):451–469
Vaughan DC, Tiku ML (2000) Estimation and hypothesis testing for non-normal bivariate distribution and applications. Math Comput Model 32:53–67
Acknowledgements
The authors are very grateful to the reviewers for their valuable comments which have substantially improved this paper.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A
1.1 A1: Elements of the Hessian Matrix H
and
1.2 A2: Elements of the Hessian Matrix \(H^{\star }\) for the Reparametrized Model
and
Appendix B
MML estimators have the following properties:
Lemma 1
For n tends to infinity, \(\hat{\mu }\) is normally distributed with mean \(\mu\) and variance \(\sigma ^2/m\).
Proof
This follows from the fact that the likelihood equation given in (7) and the modified likelihood equation given in (17) are asymptotically equivalent and \(\partial \ln L^{*}/\partial \mu\) assumes the form, see Kendall and Stuart (1961):
The normality of \(\hat{\mu}\) follows from the fact that \(E \left( \frac{\partial ^{r} \ln L^*}{\partial \mu ^{r}}\right) =0\) for all \(r \ge 3\), see Bartlett (1953).
Lemma 2
For n tends to infinity, \(\hat{\tau}_{i}\) is normally distributed with mean \(\tau _{i}\) and variance \(\sigma^2/m_{i}\).
Proof
This follows from the fact that the likelihood equation given in (8) and the modified likelihood equation given in (18) are asymptotically equivalent and \(\partial \ln L^{*}/\partial \tau _{i}\) assumes the form, see Kendall and Stuart (1961):
The normality of \(\hat{\tau }_{i}\) follows from the fact that \(E \left( \frac{\partial ^{r} \ln L^*}{\partial \tau _{i}^{r}}\right) =0\) for all \(r \ge 3\), see Bartlett (1953).
Lemma 3
Conditional on \(\mu +\tau _{i}\) being known, the distribution of \(A\hat{\sigma}^{2}/\sigma ^2\) is asymptotically Chi square with A degrees of freedom.
Proof
This follows from the fact that \(B_0/\sqrt{AC_0}\cong 0\) and \(\partial \ln L^{*}/\partial \sigma\) assumes the form, see Kendall and Stuart (1961):
where \(B_0\) and \(C_0\) are the same as B and C, respectively, with \(K_i\) is replaced by \(\mu +\tau _{i}\), see Tiku (1982) and Senoglu (2007).
Rights and permissions
About this article
Cite this article
Arslan, M.S.T., Senoglu, B. Type II Censored Samples in Experimental Design Under Jones and Faddy’s Skew t Distribution. Iran J Sci Technol Trans Sci 42, 2145–2157 (2018). https://doi.org/10.1007/s40995-017-0398-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40995-017-0398-3