Journal of Geodesy

, Volume 88, Issue 5, pp 427–439 | Cite as

An alternative method for non-negative estimation of variance components

  • Khosro Moghtased-Azar
  • Ramin Tehranchi
  • Ali Reza Amiri-Simkooei
Original Article


A typical problem of estimation principles of variance and covariance components is that they do not produce positive variances in general. This caveat is due, in particular, to a variety of reasons: (1) a badly chosen set of initial variance components, namely initial value problem (IVP), (2) low redundancy in functional model, (3) an improper stochastic model, and (4) data’s possibility of containing outliers. Accordingly, a lot of effort has been made in order to design non-negative estimates of variance components. However, the desires on non-negative and unbiased estimation can seldom be met simultaneously. Likewise, in order to search for a practical non-negative estimator, one has to give up the condition on unbiasedness, which implies that the estimator will be biased. On the other hand, unlike the variance components, the covariance components can be negative, so the methods for obtaining non-negative estimates of variance components are not applicable. This study presents an alternative method to non-negative estimation of variance components such that non-negativity of the variance components is automatically supported. The idea is based upon the use of the functions whose range is the set of all positive real numbers, namely positive-valued functions (PVFs), for unknown variance components in stochastic model instead of using variance components themselves. Using the PVF could eliminate the effect of IVP on the estimation process. This concept is reparameterized on the restricted maximum likelihood with no effect on the unbiasedness of the scheme. The numerical results show the successful estimation of non-negativity estimation of variance components (as positive values) as well as covariance components (as negative or positive values).


Variance components Non-negative Positive-valued functions Unbiased 


  1. Amiri-Simkooei AR (2007) Least-squares variance component estimation: theory and GPS applications. Ph.D. thesis, Delft University of Technology, Publication on Geodesy, 64, Netherlands Geodetic Commission, DelftGoogle Scholar
  2. Amiri-Simkooei AR, Tiberius CCJM, Teunissen PJG (2007) Assessment of noise in GPS coordinate time series: methodology and results. J Geophys Res 112(B7):2156–2202Google Scholar
  3. Brown KG (1976) Asymptotic behavior of MINQUE-type estimators of variance components. Ann Stat 4(4):746–754CrossRefGoogle Scholar
  4. Caspary WF (1987) Concepts of network and deformation analysis. Vol. 11 of monograph. School of Surveying, University of New South Wales, KensingtonGoogle Scholar
  5. Crocetto N, Gatti M, Russo P (2000) Simplified formulae for the bique estimation of variance components in disjunctive observation groups. J Geod 74:447–457CrossRefGoogle Scholar
  6. Eshagh M, Sjöberg LE (2008) The modified best quadratic unbiased non-negative estimator (MBQUNE) of variance components. Stud Geophys Geod 52(3):305–320CrossRefGoogle Scholar
  7. Förstner W (1979) Ein Verfahren zur Schätzung von Varianz- und Kovarianzkomponenten. AVN 86(11/12):446–453Google Scholar
  8. Grafarend EW (1984) Variance–covariance-component estimation of Helmert type in the Gauß–Helmert model. ZFV 109:34–44Google Scholar
  9. Grafarend EW, d’Hone A (1978) Gewichtsschätzung in geodätischen Netzen. German Geodetic Commission, MünchenGoogle Scholar
  10. Grafarend EW, Kleusberg A, Schaffrin B (1980) An introduction to the variance–covariance-component estimation of Helmert type. ZFV 105:161–180Google Scholar
  11. Grafarend EW, Schaffrin B (1993) Ausgleichungsrechnung in Linearen Modellen, German edn. BI-Wissenschaftsverlag, MannheimGoogle Scholar
  12. Grodecki J (1999) Generalized maximum-likelihood estimation of variance components with inverted gamma prior. J Geod 73(7):367–374CrossRefGoogle Scholar
  13. Grodecki J (2001) Generalized maximum-likelihood estimation of variance–covariance components with non-informative prior. J Geod 75(2–3):157–163CrossRefGoogle Scholar
  14. Groeneveld E (1994) A reparameterization to improve numerical optimization in multivariate reml (co)variance component estimation. Genet Sel Evol 26(6):537–545CrossRefGoogle Scholar
  15. Hartung J (1981) Nonnegative minimum biased invariant estimation in variance component models. Ann Stat 9(2):278–292CrossRefGoogle Scholar
  16. Harville DA (1977) Maximum likelihood approaches to variance component estimation and to related problems. J Am Stat Assoc 72(358):320–338CrossRefGoogle Scholar
  17. Hazewinkel M (2001) Encyclopaedia of mathematics. Springer, ISBN 978-1-55608-010-4Google Scholar
  18. Helmert F (1924) Die Ausgleichungsrechnung nach der Methode der kleinsten Quadrate, 3rd edn. Teubner, LeipzigGoogle Scholar
  19. Horn SD, Horn R (1975) Comparison of estimators of heteroscedastic variances in linear models. J Am Stat Assoc 70(352):872–879CrossRefGoogle Scholar
  20. Hsu R (1999) An alternative expression for the variance factors in using iterated almost unbiased estimation. J Geod 73(4):173–179CrossRefGoogle Scholar
  21. Hsu R (2001) Helmert method as equivalent of iterated almost unbiased estimation. J Surv Eng 127(3):79–89CrossRefGoogle Scholar
  22. Jennrich RI, Sampson PF (1976) Newton–Raphson and related algorithms for maximum likelihood variance component estimation. Technometrics 18(1):11–17CrossRefGoogle Scholar
  23. Koch KR (1981) Varianz- und Kovarianzkomponentenschätzung für Streckenmessungen auf Eichlinien. AVN 88(4):125–132Google Scholar
  24. Koch KR (1986) Maximum likelihood estimate of variance components. Bull Geod 60(4):329–338CrossRefGoogle Scholar
  25. Koch KR (1999) Parameter estimation and hypothesis testing in linear models. Springer, Berlin, HeidelbergGoogle Scholar
  26. Kubik K (1970) The estimation of the weights of measured quantities within the method of least squares. Bull Geod 95(1):21–40CrossRefGoogle Scholar
  27. Kusche J (2003) A Monte-Carlo technique for weight estimation in satellite geodesy. J Geod 76(11–12):641–652CrossRefGoogle Scholar
  28. LaMotte LR (1973) On non-negative quadratic unbiased estimation of variance components. J Am Stat Assoc 68(343):728–730CrossRefGoogle Scholar
  29. Nikolaidis R (2002) Observation of geodetic and seismic deformation with the global positioning system. Ph.D. thesis, University of California, San DiegoGoogle Scholar
  30. Ou Z (1989) Estimation of variance and covariance components. Bull Geod 63(2):139–148CrossRefGoogle Scholar
  31. Patterson HD, Thompson R (1971) Recovery of inter-block information when block sizes are unequal. Biometrika 58(3):545–554CrossRefGoogle Scholar
  32. Peng J, Yun S, Shuhui L, Honglei Y (2011) Minque of variance–covariance components in linear gauss–markov models. J Surv Eng 137(4):129–139CrossRefGoogle Scholar
  33. Persson CG (1981) n the estimation of variance components in linear models and related problems—with special reference to geodetic applications. Tech. rep, Royal Institute of Technology, Division of Geodesy, StockholmGoogle Scholar
  34. Pukelsheim F (1981) On the existence of unbiased non-negative estimates of variance covariance components. Ann Stat 9(2):293– 299Google Scholar
  35. Rao C, Kleffe J (1988) Estimation of variance components and applications. North-Holland, AmsterdamGoogle Scholar
  36. Rao CR (1971) Estimation of variance and covariance components—MINQUE theory. J Multivar Anal 1(3):257–275CrossRefGoogle Scholar
  37. Rao CR (1973) Linear statistical inference and its applications. Wiley, New YorkCrossRefGoogle Scholar
  38. Rao PSRS (1997) Variance components estimation: mixed models, methodologies and applications. Chapman and Hall, LondonGoogle Scholar
  39. Schaffrin B (1981a) Ausgleichung mit bedingungs-ungleichungen. AVN 88:227–238Google Scholar
  40. Schaffrin B (1981b) Best invariant covariance component estimators and its application to the generalized multivariate adjustment of heterogeneous deformation observations. Bull Geod 55:73–85CrossRefGoogle Scholar
  41. Schaffrin B (1983) Varianz-Kovarianz-Komponenten-Schätzung bei der Ausgleichung heterogener Wiederholungsmessungen. Ph.D. thesis, Rheinische Friedrich-Wilhelms-Universität, Bonn, Reihe C (Deutsche Geodätische Kommission); Heft Nr. 282.Google Scholar
  42. Searle SR, Casella G, McCulloch CE (1992) Variance components. John Wiley and Sons, New YorkCrossRefGoogle Scholar
  43. Shaw F, Geyer C (1997) Estimation and testing in constrained covariance component models. Biometrika 84(1):95–102CrossRefGoogle Scholar
  44. Sjöberg LE (1983) Unbiased estimation of variance–covariance components in condition adjustment with unknowns—a MINQUE approach. ZFV 108(9):382–387Google Scholar
  45. Sjöberg LE (1984) Non-negative variance component estimation in the Gauss–Helmert adjustment model. Manuscr Geod 9:247–280Google Scholar
  46. Sjöberg LE (1995) The best quadratic minimum bias non-negative estimator for an additive two variance component model. Manuscr Geod 20:139–144Google Scholar
  47. Sjöberg LE (2011) On the best quadratic minimum bias non-negative estimator of a two-variance component model. J Geod Sci 1(3):280–285Google Scholar
  48. Teunissen P (1988) Towards a least-squares framework for adjusting and testing of both functional and stochastic model, Tech. Rep. 26, Mathematical geodesy and positioning series, 2nd edn. Delft University of Technology, Faculty of Civil Engineering and Geosciences, Department of MGP Google Scholar
  49. Teunissen PJG, Amiri-Simkooei AR (2008) Least-squares variance component estimation. J Geod 82(2):65–82CrossRefGoogle Scholar
  50. Williams SDP, Bock Y, Fang P, Jamason P, Nikolaidis RM, Prawirodirdjo L, Miller M, Johnson DJ (2004) Error analysis of continuous GPS position time series. J Geophys Res 109(B3):B03412Google Scholar
  51. Xu P (2009) Iterative generalized cross-validation for fusing heteroscedastic data of inverse ill-posed problems. Geophys J Int 179(1):182–200CrossRefGoogle Scholar
  52. Xu P, Liu Y, Shen Y, Fukuda Y (2007) Estimability analysis of variance covariance components. J Geod 81(9):593–602CrossRefGoogle Scholar
  53. Xu P, Shen Y, Fukuda Y, Liu Y (2006) Variance component estimation in linear inverse ill-posed models. J Geod 80(2):69–81CrossRefGoogle Scholar
  54. Yu ZC (1996) A universal formula of maximum likelihood estimation of variance–covariance components. J Geod 70(4):233–240CrossRefGoogle Scholar
  55. Zhang J, Bock Y, Johnson H, Fang P, Williams S, Genrich J, Wdowinski S, Behr J (1997) Southern california permanent gps geodetic array: error analysis of daily position estimates and site velocities. J Geophys Res 102(B8):18035–18055CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • Khosro Moghtased-Azar
    • 1
  • Ramin Tehranchi
    • 2
  • Ali Reza Amiri-Simkooei
    • 3
  1. 1.Surveying Department, Faculty of Civil EngineeringTabriz UniversityTabrizIran
  2. 2.Surveying Department, Faculty of Civil EngineeringZanjan UniversityZanjanIran
  3. 3.Surveying Department, Faculty of Civil EngineeringUniversity of IsfahanIsfahanIran

Personalised recommendations