Citation: | QI Zhijun, FANG Xing, LÜ Zhipeng. Two Algorithms with High Breakdown Points Applied in Linear Regression EIV Model[J]. Geomatics and Information Science of Wuhan University, 2025, 50(1): 74-82. DOI: 10.13203/j.whugis20220441 |
Linear regression model is a basic model in the field of geodesy. To consider the structure of the coefficient matrix with the fixed column, the mixed least squares and total least squares method is implemented. However, it is easily contaminated by outliers. The M-estimator results depend on the initial value and are extremely prone to convergence badly. To increase the robustness, we propose two algorithms with high breakdown points for linear regression errors-in-variables (EIV) models, namely, the weighted total least median of squares (WTLMS) method and the weighted total least trimmed squares (WTLTS) method.
The two algorithms are extensions of traditional algorithms and use a more general stochastic model. Their breakdown points are near 50% and the two algorithms have two equivariant properties: scale equivariance and affine equivariance. The estimation formula of variance components is given. Since their objective functions are not differentiable, WTLMS and WTLTS get the solutions by the resampling algorithm and the feasible set algorithm in the EIV model respectively.
The results show that: (1) The result of the M-estimator is biased heavily from the real line, while the two proposed algorithms can obtain results close to the true value. Their performances are significantly better than M-estimator in terms of root mean square error and standard deviation. The efficiency of the two algorithms is not high, which can be further improved when the results of the two algorithms are used as the initial value of the M-estimator. The breakdown points of the two algorithms are close to 50% in the real data, which is extremely robust. (2) In the experiment of the LiDAR data, the performance of the proposed methods is better than that of the M-estimator.
The two proposed algorithms have outstanding robustness, but their complexities are high and their efficiency is not ideal. We will focus to find an easy solution with higher efficiency.
[1] |
刘经南, 曾文宪, 徐培亮. 整体最小二乘估计的研究进展[J]. 武汉大学学报(信息科学版), 2013, 38(5): 505-512.
LIU Jingnan, ZENG Wenxian, XU Peiliang. Overview of Total Least Squares Methods[J].Geomatics and Information Science of Wuhan University, 2013, 38(5): 505-512.
|
[2] |
鲁铁定, 陶本藻, 周世健. 基于整体最小二乘法的线性回归建模和解法[J]. 武汉大学学报(信息科学版), 2008, 33(5): 504-507.
LU Tieding,TAO Benzao,ZHOU Shijian.Modeling and Algorithm of Linear Regression Based on Total Least Squares[J].Geomatics and Information Scien‑ce of Wuhan University, 2008, 33(5): 504-507.
|
[3] |
WANG B, YU J, CHEN Y, et al. Efficient and Robust Solution to Universal Symmetric Transformation for 3-D Point Sets[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-15.
|
[4] |
官云兰, 刘绍堂, 周世健, 等. 基于整体最小二乘的稳健点云数据平面拟合[J]. 大地测量与地球动力学, 2011, 31(5): 80-83.
GUAN Yunlan, LIU Shaotang, ZHOU Shijian, et al. Robust Plane Fitting of Point Clouds Based on TLS[J]. Journal of Geodesy and Geodynamics, 2011, 31(5): 80-83.
|
[5] |
AMIRI-SIMKOOEI A R, JAZAERI S. Data-Snooping Procedure Applied to Errors-in-Variables Models[J]. Studia Geophysica et Geodaetica, 2013, 57(3): 426-441.
|
[6] |
WANG B, LI J C, LIU C. A Robust Weighted Total Least Squares Algorithm and Its Geodetic Applications[J]. Studia Geophysica et Geodaetica, 2016, 60(2): 177-194.
|
[7] |
龚循强, 李志林. 稳健加权总体最小二乘法[J]. 测绘学报, 2014, 43(9): 888-894.
GONG Xunqiang, LI Zhilin. A Robust Weighted Total Least Squares Method[J]. Acta Geodaetica et Cartographica Sinica, 2014, 43(9): 888-894.
|
[8] |
龚循强, 李志林. 一种利用IGGII方案的稳健混合总体最小二乘方法[J]. 武汉大学学报(信息科学版), 2014, 39(4): 462-466.
GONG Xunqiang, LI Zhilin. A Robust Mixed LS-TLS Based on IGGII Scheme[J]. Geomatics and Information Science of Wuhan University, 2014, 39(4): 462-466.
|
[9] |
刘超, 王彬, 赵兴旺, 等. 三维坐标转换的高斯-赫尔默特模型及其抗差解法[J]. 武汉大学学报(信息科学版), 2018, 43(9): 1320-1327.
LIU Chao, WANG Bin, ZHAO Xingwang, et al. Three-Dimensional Coordinate Transformation Model and Its Robust Estimation Method Under Gauss-Helmert Model[J]. Geomatics and Information Science of Wuhan University, 2018, 43(9): 1320-1327.
|
[10] |
李思达, 柳林涛, 刘志平, 等. 多变量稳健总体最小二乘平差方法[J]. 武汉大学学报(信息科学版), 2019, 44(8): 1241-1248.
LI Sida, LIU Lintao, LIU Zhiping, et al. Robust Total Least Squares Method for Multivariable EIV Model[J]. Geomatics and Information Science of Wuhan University, 2019, 44(8): 1241-1248.
|
[11] |
陶叶青, 高井祥, 姚一飞. 基于中位数法的抗差总体最小二乘估计[J]. 测绘学报, 2016, 45(3): 297-301.
TAO Yeqing, GAO Jingxiang, YAO Yifei. Solution for Robust Total Least Squares Estimation Based on Median Method[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(3): 297-301.
|
[12] |
ROUSSEEUW P J, LEROY A M. Robust Regression and Outlier Detection[M]. New Work: Wiley, 1987.
|
[13] |
FANG X, ZENG W X, ZHOU Y J, et al. On the Total Least Median of Squares Adjustment for the Pattern Recognition in Point Clouds[J]. Measurement, 2020, 160: 107794.
|
[14] |
LÜ Z P,SUI L F.The BAB Algorithm for Computing the Total Least Trimmed Squares Estimator[J]. Journal of Geodesy, 2020, 94(12): 110.
|
[15] |
ZHOU Y, FANG X. A Mixed Weighted Least Squares and Weighted Total Least Squares Adjustment Method and Its Geodetic Applications[J]. Survey Review, 2016, 48(351): 421-429.
|
[16] |
FANG X. Weighted Total Least Squares: Necessary and Sufficient Conditions, Fixed and Random Parameters[J].Journal of Geodesy,2013,87: 733-749.
|
[17] |
HAWKINS D M. The Feasible Set Algorithm for Least Median of Squares Regression[J]. Computational Statistics & Data Analysis, 1993, 16(1): 81-101.
|
[18] |
RUPPERT D. Computing S Estimators for Regression and Multivariate Location/Dispersion[J]. Journal of Computational and Graphical Statistics,1992,1(3): 253.
|
[19] |
ROUSSEEUW P J, VAN DRIESSEN K. A Fast Algorithm for the Minimum Covariance Determinant Estimator[J]. Technometrics, 1999, 41(3): 212.
|
[20] |
YANG L,SHEN Y Z,LI B F.M-Estimation Using Unbiased Median Variance Estimate[J]. Journal of Geodesy, 2019, 93(6): 911-925.
|
[21] |
刘春阳, 王坚, 王彬, 等. 基于中位参数法相关观测的抗差加权整体最小二乘算法[J]. 武汉大学学报(信息科学版), 2019, 44(3): 378-384.
LIU Chunyang, WANG Jian, WANG Bin, et al. Robust Weight Total Least Squares Algorithm of Correlated Observation Based on Median Parameter Method[J]. Geomatics and Information Science of Wuhan University, 2019, 44(3): 378-384.
|
[22] |
XU P L. Sign-Constrained Robust Least Squares, Subjective Breakdown Point and the Effect of Weights of Observations on Robustness[J]. Journal of Geodesy, 2005, 79(1): 146-159.
|
[23] |
YOHAI V J. High Breakdown-Point and High Efficiency Robust Estimates for Regression[J]. The Annals of Statistics, 1987, 15(2): 642-656.
|
[24] |
MARONNA R A, MARTIN R D, YOHAI V J. Robust Statistics[M]. New Work: Wiley, 2006.
|
[25] |
DONG Z, YANG B S, LIANG F X, et al. Hierarchical Registration of Unordered TLS Point Clouds Based on Binary Shape Context Descriptor[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 144: 61-79.
|
[26] |
苍桂华, 李明峰, 岳建平. 以入射角定权的点云数据加权总体最小二乘平面拟合研究[J].大地测量与地球动力学,2014,34(3): 95-98.
CANG Guihua, LI Mingfeng, YUE Jianping. Study on Point Clouds Plane Fitting with Weighted Total Least Squares Based on Incidence Angle Weighting[J].Journal of Geodesy and Geodynamics, 2014, 34(3): 95-98.
|