基于贝叶斯网络的重大资产重组公司预测模型

Prediction Model of Major Asset Restructured Company Based on Bayesian Network

随着资本市场的不断发展,企业重大资产重组已经成为企业优化资源配置、谋求可持续发展的重要途径。对投资者而言,在重大资产重组停牌公告发布前进行投资布局有助于获得超常收益。通过以中国A股主板2015年至2017年发布重大资产重组停牌公告的公司为样本,运用理论分析确定构建贝叶斯网络的节点,并在此基础上通过贝叶斯网络建立重大资产重组预测模型。研究发现,股权制衡度、资产减值损失占营业收入比例越大,出现控股股东变更,每股收益、每股净资产、每股营业收入越小,上市公司发生重大资产重组事件的概率越大。模型总体准确率为71.63%,表明对重大资产重组公司进行预测在很大程度上是可行的。

With the continuous development of capital market, the major asset restructuring of enterprises has become an important wayfor enterprises to optimize the allocation of resources and seek sustainable development. For investors, investment layout before major assetrestructuring announcement is helpful to get abnormal returns. By using the theoretical analysis to establish the nodes of the Bayesiannetwork, a major asset reorganization prediction model is established on the basis of the Bayesian network, which is based ona sample ofthe companies that issueda major asset restructuring stop bulletin from 2015 to 2017 of China＇s A shares. It has been found that thegreater the proportion of equity and impairment of assets is in the operating income, with the change of the controlling shareholders and thesmaller the net assets per share and the earnings per share are, the greater the probability of the major asset reorganization of the listedcompanies will have. The overall accuracy of the model is 71.63%, which indicates that it is feasible to predict major assetrestructuredcompanies.