Abstract
This paper first proves that the traditional median variance estimate is biased when the sample number is small and then proposes an unbiased median variance estimate to calibrate for the bias of the variance estimate. The scaled median variance estimate is firstly derived, and the unbiased median variance estimate is formed with independent residuals in an adjustment model no matter whether the measurements are contaminated by outliers or not. Using the unbiased median variance estimate, the M-estimate is constructed to mitigate for the biases caused by the variance estimate. The IGGIII reduction factor is used to verify the proposed algorithms by a levelling network example. Numerical analysis confirms that the proposed median variance estimate can achieve better unbiasedness for contaminated measurement set, but the dispersion of our estimate is unfortunately larger than that for the least-squares estimate.
Similar content being viewed by others
References
Baarda W (1967) Statistical concepts in geodesy. Netherlands Geodetic, Commission, Publications on Geodesy, New Series 2, No. 4, Delft. ISBN-13: 9789061322085, ISBN-10: 9061322081
Baarda W (1968) A testing procedure for use in geodetic networks. Netherland Geodetic Commission, vol 2, no 5. ISBN-13: 9789061322092, ISBN-10: 906132209X
Beaton AE, Tukey JW (1974) The fitting of power series, meaning polynomials, illustrated on band-spectroscopic data. Technometrics 16(2):147–185
Cureton EE (1968) Unbiased estimation of the standard deviation. Am Stat 22(1):22
Guttman I, Lin DKJ (1995) Robust recursive estimation for correlated observations. Stat Prob Lett 23:79–92
Hampel FR, Ronchetti EM, Rousseeuw PJ, Stahel WA (1986) Robust statistics, the approach based on influence functions. Wiley, New York
Hekimoglu S (1997) Finite sample breakdown points of outlier detection procedures. J Surv Eng 125(1):15–31
Hekimoglu S (1999) Robustifying conventional outlier detection procedures. J Surv Eng 125(2):69–86
Huber PJ (1964) Robust estimation of a location parameter. Ann Math Stat 35(2):73–101
Huber PJ (1981) Robust statistics. Wiley, New York
Huber PJ (1984) Finite sample breakdown of M- and P-estimators. Ann Stat 12:119–126
Knight NL, Wang J, Rizos C (2010) Generalized measures of reliability for multiple outliers. J Geod 84:625–635
Koch KR (1981) Deviations from the null-hypothesis to be detected by statistical tests. Bull Géodés 55:41–48
Koch KR (1999) Parameter estimation and hypothesis testing in linear models, 2nd edn. Springer, Berlin
Koch KR (2013) Robust estimation by expectation maximization algorithm. J Geod 87(2):107–116
Koch KR (2015) Minimal detectable outliers as measures of reliability. J Geod 89(5):483–490
Kok JJ (1984) On data snooping and multiple outlier testing. NOAA Technical Report, NOS NGS. 30, U.S. Department of Commerce, Rockville, Maryland
Krarup T, Kubik K, Juhl J (1980) Götterdämmerung. Over least squares. In: Proceedings of international society for photogrammetry 14th congress, Hamburg, pp 370–378
Lehmann R (2012) Improved critical values for extreme normalized and studentized residuals in Gauss–Markov models. J Geod 86(12):1137–1146
Lehmann R, Lösler M (2016) Multiple outlier detection: hypothesis tests versus model selection by information criteria. J Surv Eng 142(4):04016017
Li B, Shen Y, Lou L (2011) Efficient estimation of variance and covariance components: a case study for GPS stochastic model evaluation. IEEE Trans Geosci Remote Sens 49(1):203–210
Mao S, Wang J, Pu X (1998) Advanced mathematical statistics, 2nd edn. China Higher Education Press, Beijing
Ong EP, Spann M (1999) Robust optical flow computation based on least-median-of-squares regression. Int J Comput Vis 31(1):51–82
Pope AJ (1976) The test statistics of residuals and the detection of outliers. In: NOAA technical report NOS65 NGS 1. US Department of Commerce, National Geodetic Survey, Rockville, Maryland
Prószyñski W (2000) On outlier-hiding effects in specific Gauss–Markov models: geodetic examples. J Geod 74:581–589
Ross SM (2014) A first course in probability, 9th edn. Pearson Education Limited, Harlow
Rousseeuw P (1984) Least median of squares regression. J Am Stat Assoc 79(388):871–880
Rousseeuw PJ, Leroy AM (1987) Robust regression and outlier detection. Wiley, New York. ISBN 0-471-85233-3
Rousseeuw P, Yohai V (1984) Robust regression by means of S-estimators. In: Franke J, Härdle W, Martin D (eds) Robust and nonlinear time series analysis. Springer, Berlin, pp 256–272
Schaffrin B (1997) Reliability measures for correlated observations. J Surv Eng 123(3):126–137
Teunissen PJG (1998) Minimal detectable biases of GPS data. J Geod 72(4):236–244
Teunissen PJG (2006) Testing theory: an introduction, 2nd edn. Delft University Press, Delft. ISBN 9040719756
Van Loon JP. (2008) Robust estimation and robust re-weighting in satellite gravity modelling. In: Xu P, Liu J, Dermanis A (eds) VI Hotine-Marussi symposium on theoretical and computational geodesy. International Association of Geodesy Symposia, vol 132. Springer, Berlin, pp 43–48
Wang J, Chen Y (1994) On the reliability measure of observations. Acta Geodaetica et Cartographica Sinica, English Edition, pp 42–51
Xu P (2005) Sign-constrained robust least squares, subjective breakdown point and the effect of weights of observations on robustness. J Geod 79(1–3):146–159
Yang Y (1994) Robust estimation for dependent observations. Manuscr Geod 19:10–17
Yang Y, Cheng MK, Shum CK, Tapley BD (1999) Robust estimation of systematic errors of satellite laser range. J Geod 73:345–349
Yang Y, Song L, Xu T (2002) Robust estimator for correlated observations based on bifactor equivalent weights. J Geod 76:353–358
Yang L, Wang J, Knight NL, Shen Y (2013) Outlier separability analysis with a multiple alternative. J Geod 87:591–604
Acknowledgements
The authors are grateful to Prof. Chris Rizos for his very helpful revision on the draft of this paper. This work is sponsored by the National Key R&D Program of China (2017YFA0603103) and National Natural Science Foundation of China (41731069, 41504022).
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
The detailed proof for Eq. (12) is as follows.
Letting \( x = \frac{{\left| { \varepsilon_{1} } \right|}}{2} \) and \( y = \frac{{\left| { \varepsilon_{2} } \right|}}{2} \), from Eqs. (1) and (9); it is deduced that the corresponding PDF is
and
Letting \( z = x + y \), since variable \( x \) and \( y \) are independent, there is
Since
Substituting Eq. (74) into Eq. (73), it is deduced
Therefore, the expectation of \( z \) is
Rights and permissions
About this article
Cite this article
Yang, L., Shen, Y. & Li, B. M-estimation using unbiased median variance estimate. J Geod 93, 911–925 (2019). https://doi.org/10.1007/s00190-018-1215-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00190-018-1215-2