Measuring Corporate Credit Risk by Financial Ratios
碩士 === 世新大學 === 財務金融學研究所(含碩專班) === 93 === This thesis tests the prediction power of models assessing the probability of business financial failures. Specifically, discriminant analysis and logistic model are used to examine Taiwan’s data. Past studies have documented important variables to predict b...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2005
|
Online Access: | http://ndltd.ncl.edu.tw/handle/97538638016139532351 |
id |
ndltd-TW-093SHU05304023 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-093SHU053040232016-06-23T04:10:16Z http://ndltd.ncl.edu.tw/handle/97538638016139532351 Measuring Corporate Credit Risk by Financial Ratios 以財務比率衡量公司信用風險 Min-Hui Kuan 官旻慧 碩士 世新大學 財務金融學研究所(含碩專班) 93 This thesis tests the prediction power of models assessing the probability of business financial failures. Specifically, discriminant analysis and logistic model are used to examine Taiwan’s data. Past studies have documented important variables to predict business financial distress. The most popularly used ones are either (1) five standard financial ratios, including liquidity, profitability, leverage, solvency, and activity ratios or (2) the above five financial ratios plus the cash flow ratio. This thesis tests the performance of the two groups of variables via discriminant analysis and logistic model. Previous studies constructed their non-bankrupt firms by randomly selecting size-matched ones from the population of the same industry as that of the failed company in question. However, as the number of healthy firms of an industry is typically small in Taiwan, this procedure sometimes gets only few healthy matching firms to choose, thus resulting in selection bias. To avoid such bias I propose to choose healthy matching firms by matching size from the population of all industry. This thesis tests the performance of the two groups of variables via discriminant analysis and logistic model. In summary, for discriminant analysis and logistic model each, there are in total four models to be estimated, where two matching procedures and two groups of variables are used. The empirical results show that: (1) the Discriminant Analysis Models (Models A-D) distinguished failed from healthy firms with an accuracy rate of approximately 83.85, 84.62, 87.69, 87.69 percents, wherein the failure in one year from the date of prediction is accounted, (2) the Logistic Models (Models E-H) achieve an accuracy rate of approximately 89.23, 89.23, 90.77, and 90.77 percents respectively in doing the same job. (3) The best performance is achieved by the Logistic Model that matches healthy firms by size from all population and employs the five financial ratios. Jen-Hung Wang 王仁宏 2005 學位論文 ; thesis 80 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 世新大學 === 財務金融學研究所(含碩專班) === 93 === This thesis tests the prediction power of models assessing the probability of business financial failures. Specifically, discriminant analysis and logistic model are used to examine Taiwan’s data.
Past studies have documented important variables to predict business financial distress. The most popularly used ones are either (1) five standard financial ratios, including liquidity, profitability, leverage, solvency, and activity ratios or (2) the above five financial ratios plus the cash flow ratio. This thesis tests the performance of the two groups of variables via discriminant analysis and logistic model.
Previous studies constructed their non-bankrupt firms by randomly selecting size-matched ones from the population of the same industry as that of the failed company in question. However, as the number of healthy firms of an industry is typically small in Taiwan, this procedure sometimes gets only few healthy matching firms to choose, thus resulting in selection bias. To avoid such bias I propose to choose healthy matching firms by matching size from the population of all industry. This thesis tests the performance of the two groups of variables via discriminant analysis and logistic model. In summary, for discriminant analysis and logistic model each, there are in total four models to be estimated, where two matching procedures and two groups of variables are used.
The empirical results show that: (1) the Discriminant Analysis Models (Models A-D) distinguished failed from healthy firms with an accuracy rate of approximately 83.85, 84.62, 87.69, 87.69 percents, wherein the failure in one year from the date of prediction is accounted, (2) the Logistic Models (Models E-H) achieve an accuracy rate of approximately 89.23, 89.23, 90.77, and 90.77 percents respectively in doing the same job. (3) The best performance is achieved by the Logistic Model that matches healthy firms by size from all population and employs the five financial ratios.
|
author2 |
Jen-Hung Wang |
author_facet |
Jen-Hung Wang Min-Hui Kuan 官旻慧 |
author |
Min-Hui Kuan 官旻慧 |
spellingShingle |
Min-Hui Kuan 官旻慧 Measuring Corporate Credit Risk by Financial Ratios |
author_sort |
Min-Hui Kuan |
title |
Measuring Corporate Credit Risk by Financial Ratios |
title_short |
Measuring Corporate Credit Risk by Financial Ratios |
title_full |
Measuring Corporate Credit Risk by Financial Ratios |
title_fullStr |
Measuring Corporate Credit Risk by Financial Ratios |
title_full_unstemmed |
Measuring Corporate Credit Risk by Financial Ratios |
title_sort |
measuring corporate credit risk by financial ratios |
publishDate |
2005 |
url |
http://ndltd.ncl.edu.tw/handle/97538638016139532351 |
work_keys_str_mv |
AT minhuikuan measuringcorporatecreditriskbyfinancialratios AT guānmínhuì measuringcorporatecreditriskbyfinancialratios AT minhuikuan yǐcáiwùbǐlǜhéngliànggōngsīxìnyòngfēngxiǎn AT guānmínhuì yǐcáiwùbǐlǜhéngliànggōngsīxìnyòngfēngxiǎn |
_version_ |
1718320396968656896 |