A股上市公司董事长离职问题研究——基于Lasso模型和Logistic模型

Research on the Separation of Chairmen of A-Share Listed Companies—Based on Lasso Model and Logistic Model

摘要: 目前高管离职事件频频出现，本文选取2011~2022年A股上市公司董事长离职事件作为研究样本，将外部环境、公司经营治理和董事长个人特征作为影响因素，采用Lasso和Logistic模型进行回归分析，得到能够预测董事长离职的指标，并分析模型的预测有效性。本文研究结果表明，公司资产负债率越高，公司股权集中度越高，则董事长离职的可能性越大，当较为全面地考虑董事长离职的潜在影响因素时，Lasso模型在预测董事长离职方面的有效性要高于Logistic模型。

Abstract: At present, executive departure events occur frequently. This paper selects the 2011~2022 A-share listed company chairman departure events as a research sample, takes the external environment, the company’s operation and governance and the chairman’s personal characteristics as the influencing factors, and adopts the Lasso and Logistic model to carry on the regression analysis, to get the indexes capable of predicting the chairman’s departure, and analyzes the predictive validity of the model. The results of this paper show that the higher the company’s gearing ratio and the higher the company’s equity concentration, the higher the likelihood of the chairman’s departure, and when more comprehensively considering the potential influencing factors of the chairman’s departure, the effectiveness of the Lasso model in predicting the chairman’s departure is higher than that of the Logistic model.