演化式類神經網路在企業危機診斷上之應用-智慧資本指標的考量
Applying the Hybrid Neural Network Model to Diagnose the Enterprise Distress-Consideration of Intellectual Capital Indicator
國立台北科技大學商業自動化與管理研究所(Institute of Commerce Automation and Management, National Taipei University of Technology);國立台北科技大學商業自動化與管理研究所(Institute of Commerce Automation and Management, National Taipei University of Technology);
美國俄亥俄州立大學行銷與物流學系(Department of Marketing, and logistics The Ohio State University)
摘要
近年來,由於整體經濟環境的快速變遷,造成企業發生財務危機的可能性隨之逐年增加,因此,建立一個有效的企業危機診斷模式,是當前學術界與實務界相當重要的課題之一。本研究利用整合鑑別分析與類神經網路的兩階段建構模式方法,建構企業危機診斷分類模型。此外,在探討企業危機的衡量指標上,本研究除了參考一般傳統財務性指標外,亦加入了智慧資本指標,希望能藉由更完整多元的企業資訊,來幫助企業本身評估其自我的真實價值,並做出正確的決策。本研究經由理論與文獻的探討,建立了新的企業危機診斷模式,在經過實證的結果發現,經由鑑別分析方法針對所考量之衡量企業危機指標進行分析,得知企業經常失敗的原因,除了受到傳統財務構面指標的影響外亦受到智慧資本構面指標的影響。此外,有關整合鑑別分析與類神經網路方法所建構之企業危機診斷模式亦能確實有效的降低企業危機診斷的誤判情況,是以無論在學術研究或實務工作上,實有其相當之助益。
Abstract
In these few years, the rapid change of global economic environment has increased the occurrence possibility of financial distress. Therefore, to build up an appropriate financial distress diagnosis model has become a very important task in industry. The objective of the proposed study is to investigate the performance of enterprise distress diagnosis by integrating the artificial neural networks with discriminant analysis technique. In addition to the financial capital indicator, the intellectual capital (IC) indicator is also included in the model to measure the assets of companies. The results from the present study indicate that the proposed combined approach predict much accurate and converge much faster than that the conventional neural network approach. In the other words, a neural network takes a long time to achieve an accurate result without a good initial estimate from discriminant analysis approach. Moreover, we find out that the diagnostic correctness of enterprise distress is significantly influenced by both traditional financial indicators and IC indicators.
關鍵詞
- 類型
- 論文
- 期刊
- 臺大管理論叢
- ISSN
- 1018-1601
- 卷期
- 14卷2期
- 出版
- 2004年06月
- 頁碼
- 001–022
- 編號
- NTUEJ00010345
同期目錄
- 演化式類神經網路在企業危機診斷上之應用-智慧資本指標的考量邱志洲(Chih-Chou Chiu);簡德年(Te-Nien Chien);高淩菁(Ling-Jing Kao)・頁 001–022
- 風險值的應用與台灣共同基金績效指標之持續性王佳眞(Jai-Jen Wang);徐辜元宏(Yuan-Hong Hsu Ku)・頁 023–047
- 新金融商品開發活動之執行品質與績效-台灣銀行業及保險業之研究陳嵩(Song Chen)・頁 049–077
- 資訊系統專案團隊成員互動型態之探索汪美香(Mei-Hsiang Wang);葉桂珍(Quey-Jen Yeh)・頁 079–107
- 跳躍過程與網路股的評價-模擬準確性的探討杜化宇(Anthony H. Tu);邱志忠(Jr-Jung Chiou)・頁 109–133
- 自結盈餘與公告盈餘間差異之影響因素研究楊孟萍(Meng-Ping Yang);林瑞青(Ruey-Ching Lin);盧怡倫(I-Lun Lu)・頁 135–160
- 基金管理費之評價模式周麗娟(Li-Chuan Chou)・頁 161–178
- 漲跌幅限制變化對投資人預期之影響薛立言(L. Paul Hsueh);陳獻儀(Hsien-Yi Chen)・頁 179–196