Abstract:For a weighted total least-squares (WTLS) solution to errors-in-variables (EIV) model, three issues are essential: (1) how to acquire a perfect description of the variancecovariance matrix of the design matrix; (2) how to find an effective algorithm to deal with all the situations, particularly, when the design matrix and right hand side vector are cross-correlated; (3) how to evaluate the accuracy of the proposed methods. This paper presents a specific EIV model in which each random element of the design matrix is a function of the input vector. In such a model, the error corrupted input and output vectors are treated as the measurements equivalently. A scheme of calculating the variance-covariance matrix of the augmented design matrix based on error propagation laws is presented. It is suitable for achieving the dispersion matrices in more general cases even in the presence of the nonzero cross-covariance of dispersion matrix. It is also proved to be consistent with the rules introduced rearlier. A WTLS approach, called Newton iterative WTLS algorithm followed by an approximated accuracy assessment method is developed. Case studies demonstrate that the proposed algorithm can achieve the same accuracy as the existing WTLS or structured TLS methods. The proposed Newton WTLS algorithm aims to provide a complete WTLS adjustment method with a general weight matrix for the specific EIV model.

Optimal designs for linear models with Fredholm-type errors

A Goal-Oriented Error Estimator for the Analysis of Simplified Designs

Asymptotic Behavior for S-estimators in Random Design Linear Model with Long-Range-dependent Errors

M-estimation for linear models with exchangeable errors

Asymptotic expansion for nonparametric M-estimator in a nonlinear regression model with long-memory errors

Local Linear M‐estimation in Non‐parametric Spatial Regression

Optimal designs for linear regression models with skew-normal errors

Local linear M-estimation for spatial processes in fixed-design models

Approximate maximum likelihood estimators for linear regression with design matrix uncertainty

Optimal designs for frequentist model averaging

Optimal model averaging for partially linear models with missing response variables and error‐prone covariates

A Newton Algorithm for Weighted Total Least-Squares Solution to a Specific Errors-in-variables Model with Correlated Measurements

Doubly robust and efficient estimators for heteroscedastic partially linear single-index models allowing high dimensional covariates

Linear Model Selection when Covariates Contain Errors

Optimal Designs in Sparse Linear Models

R-optimal designs for multi-response regression models with multi-factors

On a Biased Prediction Based on Optimal Mean Square Error Criterion

Precise Error Analysis of Regularized M-estimators in High-dimensions

Optimal Model Averaging Estimation for Generalized Linear Models and Generalized Linear Mixed-Effects Models

Generalized varying coefficient partially linear measurement errors models

A reinforced learning approach to optimal design under model uncertainty