Robust Linear Regression for Undrained Shear Strength Data

  • Jun Lin
  • Guojun CaiEmail author
  • Songyu Liu
  • Anand J. Puppala
Conference paper


Outlier data has attracted considerable interesting geotechnical data. When doing classical linear least squares regression, if the regression data satisfied certain regression weights, the ordinary least squares regression is considered as the best method. However, the estimating and regression results may be inaccurate in case of these data not meeting given assumptions. Particularly in least squares regression analysis, there is some data (outliers) violating the assumption of normally distributed residuals. Under situation of regression data blending to outliers, robust regression is the best fit method. It can discriminate outliers and offer robust results when the regression data exists outliers. The purpose of this study is to make use of robust regression method to trend regression in geotechnical data analysis. Without defining absolute outliers from geotechnical testing data, outlier data of undrained shear strength is detected based on robust regression result.


Undrained shear strength Robust regression Outlier data 


  1. 1.
    Zhang, L.M., Tang, W.H., Zhang, L.L., Zheng, J.G.: Reducing uncertainty of prediction from empirical correlations. J. Geotech. Geoenviron. Eng. 130(5), 526–534 (2004)CrossRefGoogle Scholar
  2. 2.
    Ching, J., Phoon, K.: Characterizing uncertain site-specific trend function by sparse Bayesian learning. J. Eng. Mech. 143(7), 4017028 (2017)CrossRefGoogle Scholar
  3. 3.
    Baecher, G.B., Christian, J.T.: Reliability and Statistics in Geotechnical Engineering. Wiley, New York (2003)Google Scholar
  4. 4.
    Gillins, D.T., Bartlett, S.F.: Multilinear regression equations for predicting lateral spread displacement from soil type and cone penetration test data. J. Geotech. Geoenviron. Eng. 141(4), 04013047 (2015)CrossRefGoogle Scholar
  5. 5.
    Yuen, K.V., Ortiz, G.A.: Outlier detection and robust regression for correlated data. Comput. Methods Appl. Mech. Eng. 313, 632–646 (2016)CrossRefGoogle Scholar
  6. 6.
    Gürünlü Alma, Ö.: Comparison of robust regression methods in linear regression. Int. J. Contemp. Math. Sci. 6(9–12), 409–421 (2011)Google Scholar
  7. 7.
    Davies, P.L.: Aspects of robust linear regression. Ann. Stat. 21(4), 1843–1899 (1993)CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  • Jun Lin
    • 1
  • Guojun Cai
    • 1
    Email author
  • Songyu Liu
    • 1
  • Anand J. Puppala
    • 2
  1. 1.Institute of Geotechnical EngineeringSoutheast UniversityNanjingChina
  2. 2.Department of Civil EngineeringThe University of Texas at ArlingtonArlingtonUSA

Personalised recommendations