Corrected score estimators in linear multivariate multiple regressionmodels with heteroscedastic measurement errors | |
รหัสดีโอไอ | |
Creator | 1. Wannaporn Junthopas 2. Jirawan Jitthavech 3. Vichit Lorchirachoonkul |
Title | Corrected score estimators in linear multivariate multiple regressionmodels with heteroscedastic measurement errors |
Publisher | Research and Development Office, Prince of Songkla University |
Publication Year | 2018 |
Journal Title | Songklanakarin Journal of Science and Technology |
Journal Vol. | 40 |
Journal No. | 3 |
Page no. | 506 |
Keyword | corrected score, grouped heteroscedasticity, heteroscedastic measurement errors, repeated measurements |
URL Website | http://rdo.psu.ac.th/sjstweb/index.php |
ISSN | 0125-3395 |
Abstract | In this study, the knowledge of estimation theory based on the corrected score (CS) approach is extended in a linearmultivariate multiple regression model with heteroscedastic measurement errors (HMEs) and an unknown HME variance. Theheteroscedasticity of the HME variance is assumed to be capable of being grouped into similar patterns and can be estimated bythe pooled variance of the observations of the variable with HME in repeated measurements. The statistical properties of theproposed CS estimator are analytically investigated. The simulation results confirm the theoretical results. The proposed CSestimator outperforms the OLS estimator under all simulation conditions. |