Title: Robust estimation and confidence interval in meta-regression models
Authors: Dalei Yu - Yunnan University of Finance and Economics (China) [presenting]
Chang Ding - Yunnan University of Finance and Economics (China)
Na He - Industry and Commerce Administration of Yunnan Province (China)
Ruiwu Wang - Northwestern Polytechnical University (China)
Xiao-Hua Zhou - University of Washington (United States)
Lei Shi - Yunnan University of Finance and Economics (China)
Abstract: Meta-analysis provides a quantitative method for combining results from independent studies with the same treatment. However, existing estimation methods are sensitive to the presence of outliers in the datasets. We study the robust estimation for the random effects meta-regression. Hubers rho function and Tukeys biweight function are adopted to derive the formulae of robust maximum likelihood estimators. The corresponding algorithms are developed. The asymptotic confidence interval and second-order-corrected confidence interval are investigated. Simulation studies show that the robust estimators are promising and outperform the conventional maximum likelihood and restricted maximum likelihood estimators when outliers exist in the dataset. Results in case studies further support the eligibility of our methods in practical situations.