再探公司治理對經營績效與財務報導品質之影響：因素分析與類神經網路之應用

Further Evidence on the Effect of Corporate Governance on Firm Performance and Financial Reporting Quality: The Application of Factor Analysis and Neural Network

過去有關公司治理之研究對「公司治理」此一複雜結構型變數之衡量相當分歧，據此，本研究統整國內於2002 年至2009 年間發表於TSSCI 期刊之文獻，自38 篇相關研究中歸納出26 個公司治理變數，利用因素分析法粹取出9 個構面，再以類神經網路模型測試這9 個公司治理因素對公司績效與財務報導品質之預測能力。研究結果發現，各因素在解釋經營績效與盈餘品質方面確實有顯著不同的重要性。本研究結果可提供後續學術研究一個衡量公司治理變數之架構，亦有助於協助企業界利用有限的資源建立完善之公司治理制度，並可作為政府單位發展公司治理評鑑系統之參考。

The empirical research on corporate governance has not produced a consistent and comparable set of results, partially because of the lack of reliable and valid measures for the complex construct termed “corporate governance.” We survey 38 studies published in TSSCI journals during 2002 to 2009 and summarize 26 structural measures of corporate governance. An exploratory principal component analysis suggests that there are 9 dimensions. Using neural network model, we find that these factors, despite of high predictive ability, have differential effect on firm performance and earnings quality. Our results provide implications to corporations and governmental agencies who are interested in developing a comprehensive corporate governance system.