Previous |  Up |  Next

Article

Keywords:
unbalanced mixed linear model; variance components; Wald test; ANOVA-like test; Bartlett-Scheffé tests
Summary:
The paper presents some approximate and exact tests for testing variance components in general unbalanced mixed linear model. It extends the results presented by Seifert (1992) with emphasis on the computational aspects of the problem.
References:
[1] R. B. Davies: The distribution of a linear combination of chi-square random variables. J. Roy. Statist. Soc. Ser. C 29 (1980), 323–333.
[2] S. Gnot, M. Jankowiak-Rosłanowska and A. Michalski: Testing for hypothesis in mixed linear models with two variance components. Listy Biometryczne—Biometrical Letters 29 (1992), 13–31.
[3] S. Gnot, A.  Michalski: Tests based on admissible estimators in two variance components models. Statistics 25 (1994), 213–223. DOI 10.1080/02331889408802446 | MR 1366826
[4] J. P. Imhof: Computing the distribution of quadratic forms in normal variables. Biometrics 48 (1961), 419–426. DOI 10.1093/biomet/48.3-4.419 | MR 0137199 | Zbl 0136.41103
[5] A. I. Khuri, T.  Mathew and B. K. Sinha: Statistical Tests for Mixed Linear Models. J. Wiley, New York, 1998. MR 1601351
[6] J. Kleffe, B. Seifert: On the role of MINQUE in testing of hypotheses under mixed linear models. Comm. Statist. Theory Methods 17 (1988), 1287–1309. DOI 10.1080/03610928808829680 | MR 0942977
[7] L. R. LaMotte, A. McWhorter and R. A. Prasad: Confidence intervals and tests on the ratio in random models with two variance components. Comm. Statist. Theory Methods 17 (1988), 1135–1164. DOI 10.1080/03610928808829675 | MR 0942972
[8] T. H. Lin, D.  Harville: Some alternatives to Wald’s confidence interval and test. J. Amer. Statist. Assoc. 86 (1991), 179–187. DOI 10.1080/01621459.1991.10475018 | MR 1137109
[9] T. Mathew: Optimum invariant tests in mixed linear models with two variance components. In: Statistical Data Analysis and Inference (Y. Dodge, ed.), North-Holland, Amsterdam, 1989, pp. 381–388. MR 1089650 | Zbl 0735.62068
[10] A. Michalski, R.  Zmyślony: Testing hypotheses for variance components in mixed linear models. Statistics 27 (1996), 297–310. DOI 10.1080/02331889708802533 | MR 1375446
[11] T. Mathew, B. K.  Sinha: Optimum tests in unbalanced two-way models without interactions. Ann. Statist. 16 (1988), 1727–1740. DOI 10.1214/aos/1176351065 | MR 0964950
[12] A. Olsen, J. Seely and D. Birkes: Invariant quadratic estimation for two variance components. Ann. Statist. 4 (1976), 878–890. DOI 10.1214/aos/1176343586 | MR 0418345
[13] C. R. Rao: Estimation of variance components—MINQUE theory. J. Multivariate Anal. 1 (1971), 257–275. DOI 10.1016/0047-259X(71)90001-7 | MR 0301869
[14] C. R. Rao, J.  Kleffe: Estimation of Variance Components and Applications. North-Holland Publishing Company, Amsterdam, 1988. MR 0933559
[15] J. F. Seely, Y. El-Bassiouni: Applying Wald’s variance component test. Ann. Statist. 11 (1983), 197–201. MR 0684876
[16] B. Seifert: Estimation and test of variance components using the MINQUE-method. Statistics 16 (1985), 621–635. DOI 10.1080/02331888508801898 | MR 0803502 | Zbl 0593.62066
[17] B. Seifert: Exact tests in unbalanced mixed analysis of variance. J. Statist. Plann. Inference 30 (1992), 257–266. DOI 10.1016/0378-3758(92)90086-8 | MR 1157765 | Zbl 0746.62016
[18] A. Wald: A note on the analysis of variance with unequal class frequencies. Ann. Statist. 11 (1940), 96–100. DOI 10.1214/aoms/1177731946 | MR 0001502
[19] A. Wald: A note on regression analysis. Ann. Statist. 18 (1947), 586–589. DOI 10.1214/aoms/1177730350 | MR 0023498 | Zbl 0029.30703
[20] P. H. Westfall: Power comparisons for invariant variance ratio tests in mixed ANOVA models. Ann. Statist. 17 (1989), 318–326. DOI 10.1214/aos/1176347019 | MR 0981453 | Zbl 0669.62055
[21] V. Witkovský: Optimality of the ANOVA-like test in model with two variance components. In: MEASUREMENT 99. Proceedings of the International Conference on Measurement, Smolenice, Slovak Republic, 26–29 April 1999, I. Frollo, A.  Plačková (eds.), 1999, pp. 28–31.
Partner of
EuDML logo