Linear Regression Models under Conditional Independence Restrictions
A double sampling design is considered where the first sample consists of observations of \((Y,Z)\) and the second one consists of \((Z,X)\). The aim is to estimate parameters of the regression of \(Y\) on \(X\). The regression models are linear and Gaussian. The main assumption is the conditional independence: \(L(X,Y\mid Z)=L(Y\mid X)L(X\mid Z)\). The authors derive the ML estimator for the regression coefficients and error variance. Unbiasedness of the coefficients estimator and its variance are evaluated. Asymptotic normality is demonstrated. Results of agriculture data analysis are presented.
- A Unified Approach to Regression Analysis Under Double-Sampling Designs
- Conditional independence models for seemingly unrelated regressions with incomplete data
- Regression analysis with doubly censored data
- Efficient estimation in models with independence restrictions
- Lattice conditional independence models for seemingly unrelated regressions
This page was built for publication: Linear Regression Models under Conditional Independence Restrictions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4455968)