碩士 / 國立彰化師範大學 / 會計學系 / 96 / Before the outbreak of financial crisis, there are always some premonitions on the financial statement indicating that the operation or finance of the company has already shown signs of poor management or cash-flow stalemate. This study attempts to use the ratios between finance and non-finance of the proclamation of financial statements to construct an effective financial distress prediction model in an attempt to help the investors and creditors inspect the qualities of their decision policies and to make the decision policies faster and more effective.
This study uses logistic regression, CART and C5.0 in decision tree classification to construct financial distress prediction model. The findings of this study are as follows:
1. There is no way to increase the predictive ability of financial distress prediction model by means of factor analysis selective research variables.
2. With the three pre-warning models, using C5.0 to construct financial distress prediction model has the highest overall predictive accuracy.
3. The predictive ability constructed by financial distress prediction model by means of decision tree classification is superior to that of logistic regression.
4. It is discovered that by observation of time sequence, the two seasons before the happening of financial crisis show the best predictive rates.
| Identifer | oai:union.ndltd.org:TW/096NCUE5385003 |
| Date | January 2008 |
| Creators | 沙昱宏 |
| Contributors | 陳光谷 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 96 |
Page generated in 0.0114 seconds