Robustness of Quasi-Likelihood Estimating Equations for Model Misspecification
碩士 === 國立中正大學 === 統計科學所 === 96 === In this paper, we use the concept of QL function together with a simple working covariance matrix todevelop an estimating equation for spatially correlated count data.Even with a misspecified covariance structure in the estimating equation, we show that the propose...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2008
|
Online Access: | http://ndltd.ncl.edu.tw/handle/12149060457258270353 |
Summary: | 碩士 === 國立中正大學 === 統計科學所 === 96 === In this paper, we use the concept of QL function together with a simple working covariance matrix todevelop an estimating equation for spatially correlated count data.Even with a misspecified covariance structure in the estimating equation, we show that the proposedestimate may still be consistent and have asymptotic normality under some regularity cinditions. Wealso conduct simulations to compare the performance of the proposed method QLwand the estimation method QLmproposed by McShane et al. (1997).
|
---|