Least absolute deviations estimation for uncertain regression with imprecise observations

Traditionally regression analysis answers questions about the relationships among variables based on the assumption that the observation values of variables are precise numbers. It has long been dominated by least squares, mostly due to the elegant theoretical foundation and ease of implementation. However, in many cases, we can only get imprecise observation values and the assumptions upon which the least squares is based may not be valid. So this paper characterizes the imprecise data in terms of uncertain variables and proposes a novel robust approach under the principle of least absolute deviations to estimate the unknown parameters in uncertain regression models. Furthermore, some general estimate approaches are also explored. Finally, numerical examples illustrate that our estimate is more robust than the least squares implying it is more suitable to handle observations with outliers.

[1]  Abraham Charnes,et al.  Optimal Estimation of Executive Compensation by Linear Programming , 1955 .

[2]  Baoding Liu,et al.  Uncertainty Theory - A Branch of Mathematics for Modeling Human Uncertainty , 2011, Studies in Computational Intelligence.

[3]  Yi Yang,et al.  Some new ranking criteria in data envelopment analysis under uncertain environment , 2017, Comput. Ind. Eng..

[4]  Edward L. Frome,et al.  A revised simplex algorithm for the absolute deviation curve fitting problem , 1979 .

[5]  Baoding Liu,et al.  Uncertainty Theory - A Branch of Mathematics for Modeling Human Uncertainty , 2011, Studies in Computational Intelligence.

[6]  David Birkes,et al.  Alternative Methods of Regression: Birkes/Alternative , 1993 .

[7]  Peter J. Huber,et al.  The place of the L1-norm in robust estimation , 1987 .

[8]  F. Yates.  Syllabus for Professional Statisticians who Intend to Become Experts in Sampling Census and Survey Methods , 1950 .

[9]  Baoding Liu,et al.  Uncertain time series analysis with imprecise observations , 2018, Fuzzy Optimization and Decision Making.

[10]  I. Barrodale,et al.  An Improved Algorithm for Discrete $l_1 $ Linear Approximation , 1973 .

[11]  Baoding Liu,et al.  Uncertain data envelopment analysis with imprecisely observed inputs and outputs , 2018, Fuzzy Optim. Decis. Mak..

[12]  Neal C. Gallagher,et al.  On Edgeworth's method for minimum absolute error linear regression , 1994, IEEE Trans. Signal Process..

[13]  Baoding Liu.  Some Research Problems in Uncertainty Theory , 2009 .

[14]  Yin Zhang.  Primal-dual interior point approach for computingl1-solutions andl∞-solutions of overdetermined linear systems , 1993 .

[15]  Baoding Liu,et al.  Residual and confidence interval for uncertain regression model with imprecise observations , 2018, J. Intell. Fuzzy Syst..

[16]  Baoding Liu.  Why is There a Need for Uncertainty Theory , 2012 .

[17]  Baoding Liu,et al.  Uncertain regression analysis: an approach for imprecise observations , 2018, Soft Comput..

[18]  Alireza Ghaffari Hadigheh,et al.  A novel DEA model based on uncertainty theory , 2018, Ann. Oper. Res..

[19]  Kai Yao.  Uncertain Statistical Inference Models With Imprecise Observations , 2018, IEEE Transactions on Fuzzy Systems.

[20]  D. W. Scott,et al.  The L 1 Method for Robust Nonparametric Regression , 1994 .

[21]  G. O. Wesolowsky.  A new descent algorithm for the least absolute value regression problem , 1981 .

[22]  Joyce Snell,et al.  6. Alternative Methods of Regression , 1996 .