Advances and Applications in Statistics
Volume 49, Issue 2, Pages 137 - 166
(August 2016) http://dx.doi.org/10.17654/AS049020137 |
|
ESTIMATION IN FUNCTIONAL MEASUREMENT ERROR MODELS WITH COMPOUND SYMMETRIC COVARIANCE STRUCTURED EQUATION ERRORS
B. Bhargavarama Sarma and B. Shobha
|
Abstract: Measurement error models with normally and independently distributed equation errors are studied extensively in literature. In this paper, we study the linear functional measurement error models with normally distributed equation errors that are not independent but have compound symmetric covariance structure. We assume the identifiability condition that variance in measurement error is known. We derive estimators of the coefficients when the intraclass correlation is known and also when it is not known. We prove that both the estimators are consistent and also discuss the other large sample properties. Simulation work is done to support our results. |
Keywords and phrases: measurement error, consistent estimation, asymptotic properties, compound symmetric. |
|
Number of Downloads: 434 | Number of Views: 1267 |
|