10.6084/M9.FIGSHARE.14499620.V1
Dongyu Li
Dongyu
Li
Lei Wang
Lei
Wang
Improved <i>k</i>th power expectile regression with nonignorable dropouts
<p>The <i>k</i>th (<math><mn>1</mn><mo><</mo><mi>k</mi><mo>≤</mo> <mn>2</mn></math>) power expectile regression (ER) can balance robustness and effectiveness between the ordinary quantile regression and ER simultaneously. Motivated by a longitudinal ACTG 193A data with nonignorable dropouts, we propose a two-stage estimation procedure and statistical inference methods based on the <i>k</i>th power ER and empirical likelihood to accommodate both the within-subject correlations and nonignorable dropouts. Firstly, we construct the bias-corrected generalized estimating equations by combining the <i>k</i>th power ER and inverse probability weighting approaches. Subsequently, the generalized method of moments is utilized to estimate the parameters in the nonignorable dropout propensity based on sufficient instrumental estimating equations. Secondly, in order to incorporate the within-subject correlations under an informative working correlation structure, we borrow the idea of quadratic inference function to obtain the improved empirical likelihood procedures. The asymptotic properties of the corresponding estimators and their confidence regions are derived. The finite-sample performance of the proposed estimators is studied through simulation and an application to the ACTG 193A data is also presented.</p>
Medicine
Genetics
Neuroscience
Biotechnology
Mathematical Sciences not elsewhere classified
Computational Biology
Taylor & Francis
2021
2021-04-28
2024-02-19
Journal contribution
1158276 Bytes
10.6084/m9.figshare.14499620
10.1080/02664763.2021.1919606
CC BY 4.0