Robust estimation of the parametric regression model under semi-supervised transfer learning

In the era of big data, we often encounter the challenges of missing data and insufficient sample size, which will reduce the accuracy of analysis results and the reliability of models. To overcome these difficulties, this paper introduces a novel robust estimation method based on a parametric regression model. This method combines the advantages of semi-supervised learning and transfer learning, and integrates additional data information through non-parametric imputation techniques and density ratio functions, aiming to reduce the errors that may be generated by the model. Through theoretical analysis, we prove the asymptotic normality of the proposed method and compare the variance performance of different estimation methods. Finally, through a series of simulation experiments and real data tests, we verify the effectiveness and superiority of the new method.