none

• Main Authors: Mei-Ching Huang, 黃梅菁
• Other Authors: Torng-Her Lee
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/x85ftf

碩士 === 國立東華大學 === 國際經濟研究所 === 96 === In contrast to the traditional model of financial distress construction in the firm level using only financial ratio variables in financial statements, this paper uses earnings management index, and accounting variables to construct models for business financial distress. We adopted 47 companies of financial distress and 128 healthy companies to be our samples during the period 1996 to 2005. For matching is used in the analysis, we construct 7 warning models for financial distress, and then use the logistic regression to examine the effects of earnings management index, and financial ratio variables on the predictive power of financial distress. The empirical results show that, for earnings management index, the enterprise is more paid attention to the liquidity doesn't cut quantity and should accrued item and non- liquidity don't cut quantity and should accrued item, cut quantity should accrued item to the finance crisis of the probability present negative influence and indicate that the operation of the management authorities cuts quantity and should accrued item to make a company's debt better up the surplus, the operation degree of the surplus management and the dissimilarity of way, make the influence presented a contrary phenomenon, the probability of the occurrence of the finance crisis is also more big. For financial ratio variables, the debt ratio has significantly positive effect on financial distress probability. The control variable (company scale) is taking place a finance crisis for that year and the continuous year before attain the negative influence of notable level and indicate a company the scale is more big and the probability of the occurrence of the finance crisis is also more small. The financial early warning model constructed using earnings management index and financial variables has the highest predictive power of accuracy. After representing the model of the surplus management to join all finance ratio of to the exaltation of the explanation ability that depends on variable, the past surplus management variable joins the variable of the finance ratio to have better classification accuracy. This finding has a substantial implication and contribution to the financial warning modeling of corporate distress.