Research on the Effectiveness of Deep Learning−Based Agency Cost Suppression Strategy: A Case Study of State−Owned Enterprises in Mainland China

The mixed ownership reform aims to improve the property rights structure of state−owned enterprises (SOEs) and reduce agency costs, current strategies mainly include equity blending by introducing external non−state capital, executive assignments, employee stock ownership. In this paper, 953 valid data A−shares listed in Shanghai Shenzhen from 2008 2020 are used as samples construct indicators strategy literature statistics method. After obtaining multiple impact indicators, regression model corporate cost suppression is constructed MATLAB software using a machine learning algorithm. On basis, performance algorithms compared, it found that integrated optimization−based bag−boosting study effect hybrid costs SOEs, proportional setting when optimal also explored. Finally, laws different influencing factors on explored separately eigenvalue results show proportion shareholding first largest shareholder sin−functional with SOEs majority shareholders introduced into SOEs’ mix, tend decrease after become privately held enterprises. greater number supervisors appointed shareholders, supervisory restraint SOE managers better suppressing costs. participation non−state−owned company’s business decisions executives special resource advantages intensify occurrence self−interest increase SOEs. implementation an plan plays role supervision managers, which reduces Based this, can provide support for government policy promote process layer layer, theoretical reference deepen incentivize shareholding, empower govern thus