Linear models and convex geometry: aspects of non-negative variance estimationFootnote¹

¹Extended version of a paper presented at the VI International Conference on Mathe matical Statistics, Wisla, Poland, December 7-13, 1978.

References

• Aitken , A.C. 1935 . On least squares and linear combination of observations . Proc. Roy. Soc. Edinburgh , 55 : 42 – 48 .
   Google Scholar
• Albert , A. 1973 . The Gauss-Markov theorem for regression models with possibly singular covariances . SIAM J. Appl. Math , 24 : 182 – 187 .
   Web of Science ®Google Scholar
• Anderson , R.L. 1965 . Query 9. Negative variance estimates . Technometrics , 7 : 75 – 76 .
   Web of Science ®Google Scholar
• Atiqullah , M. 1962 . The estimation of residual variance in quadratically balanced least-squares problems and the robustness of the F-test . Biometrika , 49 : 83 – 91 .
   Web of Science ®Google Scholar
• Baksalary , J.K. and Kala , R. 1978 . Relationships between some representations of the best linear unbiased estimator in the general Gauss-Markoff model . SIAM J. Appl. Math , 35 : 515 – 520 .
   Web of Science ®Google Scholar
• Bellman , R. and Fan , K. On systems of linear inequalities in Hermitian matrix variables . Proceedings of Symposia in Pure Mathematics . Edited by: Klee , V. Vol. VII , Providence, R.I : American Mathematical Society .
   Google Scholar
• Borwein , J.M. and Wolkowicz , H. 1979 . Characterizations of optimality for the abstract convex program Halifax, Nova Scotia Typescript
   Google Scholar
• Brown , K.G. 1977 . On estimation of diagonal covariance matrices by MINQUE . Comm. Statist. A - Theory Methods , 6 : 471 – 484 .
   Web of Science ®Google Scholar
• Brown , K.G. 1978 . Estimation of variance components using residuals . J. Amer.Statist. Assoc , 73 : 141 – 146 .
   Web of Science ®Google Scholar
• Drygas , H. 1972 . The estimation of residual variance in regression analysis . Math. Operationsforsch. Statist , 3 : 373 – 388 .
   Google Scholar
• Drygas , H. and Hupet , G. 1979 . Non-negativeness of the best quadratic unbiased estimator of the variance in the linear regression model preprint Nr. 18/1979. Gesamt- hochschule Kassel
   Google Scholar
• Federer , W.T. 1968 . Non-negative estimators, for components of variance . J. Roy, Statist, Soc. C , 17 : 171 – 174 .
   Google Scholar
• Fuller , W.A. and Rao , J.N.K. 1978 . Estimation for a linear regression model with unknown diagonal covariance matrix . Ann. Statist , 6 : 1149 – 1158 .
   Web of Science ®Google Scholar
• Judge , G.G. and Takayama , T. 1966 . Inequality restrictions in regression analysis . J. Amer Statist. Assoc , 61 : 166 – 181 .
   Web of Science ®Google Scholar
• Hartley , H.O. and Rao , J.N.K. 1967 . Maximum likelihood estimation for the mixed analysis of variance model . Biometrika , 54 : 93 – 108 .
   PubMed Web of Science ®Google Scholar
• Hartung , J. 1981 . Non-negative minimum biased invariant estimation in variance component models . Ann. Statist , 9 ( 2 )
   Web of Science ®Google Scholar
• Harville , D.A. 1977 . Maximum likelihood approaches to variance component estimation and to related problems . J. Amer. Statist. Assoc , 72 ( 2 ) : 320 – 340 .
   Web of Science ®Google Scholar
• Hemmerle , W.J. and Hartley , H O. 1973 . Computing maximum likelihood estimates for the mixed A. O. V. model using the W transformation . Technometrics , 15 ( 2 ) : 819 – 831 .
   Web of Science ®Google Scholar
• Herbach , L.H. 1959 . Properties of model II — type analysis of variance tests, A: Optimum nature of the F-test for model II in the balanced case . Ann, Math. Statist , 30 ( 2 ) : 939 – 959 .
   Google Scholar
• Horn , S.D. and Horn , R.A. 1975 . Comparison of estimators of heteroscedastic variances in linear models . J. Amer. Statist. Assoc , 70 ( 2 ) : 872 – 879 .
   Web of Science ®Google Scholar
• Hudson , D.J. 1969 . Least-squares fitting of a polynomial constrained to be either non-negative, non-decreasing or convex . J. Roy. Statist. Soc. B , 31 ( 2 ) : 113 – 118 .
   Google Scholar
• Kleffe , J. 1977 . Optimal estimation of variance components — a survey . Sankhyā B , 39 ( 2 ) : 211 – 244 .
   Google Scholar
• Kolmogorov , A.N. 1946 . On the proof of the method of least squares . Uspehi Mat. Nauk , 1 ( 11 ) : 57 – 70 . in Russian
   Google Scholar
• Kozek , A. 1980 . Personal communication
   Google Scholar
• Krafft , O. 1970 . “ Programming methods in statistics and probability theory ” . In Nonlinear Programming , Edited by: Rosen , J.B. , Mangasarian , O.L. and Ritter , K. New York : Academic Press .
   Google Scholar
• LaMotte , L.R. 1973 . On non-negative quadratic unbiased estimation of variance components . J. Amer. Statist. Assoc , 68 : 728 – 730 .
   Web of Science ®Google Scholar
• Liew , C.K. 1976 . Inequality constrained least-squares estimation . J. Amer. Statist. Assoc , 71 : 746 – 751 .
   Web of Science ®Google Scholar
• McHugh , R.B. and Mielke , P.W. Jr . 1968 . Negative variance estimates and statistical dependence in nested sampling . J. Amer. Statist. Assoc , 63 : 1000 – 1003 .
   Web of Science ®Google Scholar
• Nelder , J.A. 1954 . The interpretation of negative components of variance . Biometrika , 41 : 544 – 548 .
   Web of Science ®Google Scholar
• Pukelsheim , F. 1974 . “ Schätzen von Mittelwert und Streuungsmatrix in Gauss- Markoff-Modeilen ” . In Diplomarbeit , Universitat Freiburg im Breisgau . Institut für Mathematische Stochastik
   Google Scholar
• Pukelsheim , F. 1976 . Estimating variance components in linear models . J. Multi- variate Anal , 6 : 626 – 629 .
   Web of Science ®Google Scholar
• Pukelsheim , F. 1977a . “ Punktschätzungen in linearen statistischen Modellen ” . In Dissertation , Universität Freiburg im Breisgau . Institut für Mathematische Stochastik
   Google Scholar
• Pukelsheim , F. 1977b . Linear models and convex programs: unbiased nonnegative estimation in variance component models , Stanford University . Technical Report No. 104. Department of Statistics
   Google Scholar
• Pukelsheim , F. 1978a . Examples for unbiased nonnegative estimation in variance component models , Stanford University . Technical Report No. 113. Department of Statistics
   Google Scholar
• Pukelsheim , F. 1978b . A quick introduction to mathematical programming with applications to most powerful tests, nonnegative variance estimation, and optimal design theory , Stanford University . Technical Report No. 128. Department of Statistics
   Google Scholar
• Pukelsheim , F. 1979 . Classes of linear models 69 – 83 . [53]
   Google Scholar
• Pukelsheim , F. 1981 . On the existence of unbiased non-negative estimates of variance covariance components . Ann. Statist , 9 ( 2 )
   Web of Science ®Google Scholar
• Pukelsheim , F. and Styan , G.P.H. 1979 . Nonnegative definiteness of the estimated dispersion matrix in a multivariate linear model . Bull. Acad. Polon. Sci. Sér. Sci. Math , 27 ( 2 ) : 327 – 330 .
   Google Scholar
• Rao , C.R. Least squares theory using an estimated dispersion matrix and its application to measurement of signals . Proc. Fifth Berkeley Symp. Math, Statist, Prob . Vol. 1 , pp. 355 – 372 .
   Google Scholar
• Rao , C.R. 1973 . Linear Statistical Inference and Its Applications , New York : John Wiley & sons .
   Google Scholar
• Rao , C.R. 1978 . Choice of best linear estimators in the Gauss-Markoff model with a singular dispersion matrix . Comm. Statist. A — Theory Methods , 7 : 1199 – 1208 .
   Web of Science ®Google Scholar
• Rao , J.N.K. 1973 . On the estimation of heteroscedastic variances . Biometrics , 29 : 11 – 24 .
   Web of Science ®Google Scholar
• Rao , P.S.R.S and Chaubey , Y.P. 1978 . Three modifications of the principle of the MINQUE . Comm. Statist. A – Theory Methods , 7 : 767 – 778 .
   Web of Science ®Google Scholar
• Rich , D.K. and Brown , K.G. 1979 . Estimation of variance components using residuals: some empirical evidence 139 – 154 . [53]
   Google Scholar
• Rockafellar , R.T. 1970 . Convex Analysis , Princeton, N.J : Princeton University Press .
   Google Scholar
• Sahai , H. 1974 . On negative estimates of variance components under finite population models . South African Statist. J , 8 : 157 – 166 .
   Web of Science ®Google Scholar
• Sahai , H. 1979 . A bibliography on variance components . Internat. Statist. Rev , 47 : 177 – 222 .
   Web of Science ®Google Scholar
• Schenné , H. 1959 . The Analysis of Variance , New York : John Wiley & Sons .
   Google Scholar
• Searle , S.R. 1971 . Linear Models , New York : John Wiley & Sons .
   Google Scholar
• Seely , J. 1970 . Linear spaces and unbiased estimation – application to the mixed model . Ann. Math. Statist , 41 : 1735 – 1748 .
   Google Scholar
• Seely , J. 1971 . Quadratic subspaces and completeness . Ann. Math. Statist , 42 : 710 – 721 .
   Google Scholar
• Srivastava , J.N. 1975 . A Survey of Statistical Design and Linear Models , Amsterdam : North Holland Publishing Company .
   Google Scholar
• Van Vleck , L.D. Variance Components and Animal Breeding . Proceedings of a Conference in Honor of C. B. Henderson . Edited by: Searle , S.R. Ithaca, N. Y : Cornell University .
   Google Scholar
• Verdooren , L.R. 1980 . On estimation of variance components Statist . Neerlandica , 34 : 83 – 106 .
   Google Scholar
• Witting , H. 1966 . Mathemaiische Statistik , Stuttgart : B. G. Teubner .
   Google Scholar
• Zyskind , G. 1967 . On canonical forms, non-negative covariance matrices and best and simple least squares linear estimators in linear models . Ann. Math. Statist , 38 : 1092 – 1109 .
   Google Scholar
• Zyskind , G. 1975 . Error structures, projections and conditional inverses in linear model theory 647 – 663 . [52]
   Google Scholar