Multi-parameter generalized smooth adjustment estimation in high dimensional correlation data

In this paper, we propose a multi-parameter generalized adaptive smooth adjustment correlation effect (MGASACE) estimator for high-dimensional correlation data by adding a parameter to the generalized adaptive smooth adjustment for correlated effects estimator (GASACE) presented by Ming, Chen and Yang (2023). We apply random cross-verification process to fix these parameters, reducing dramatically the computational cost. Additionally, we prove the oracle properties of our MGASACE method under some assumption conditions. Finally, the effectiveness of our algorithm in estimation, prediction, and model accuracy has been verified through numerical simulation experiments and testing on breast cancer data.