Publication |
2011.
|
Summary/Abstract |
Using non-probability purposive sampling, this study matched 23 Taiwanese OTC (Over-the-Counter) bankrupt corporations with 23 non-bankrupt corporations during the period 1999-2005. On the basis of the sample, a predictive and secondary research design was conducted to construct a rolling-prediction model and validate its prediction in Taiwan on OTC corporation bankruptcy risks. The empirical results indicated that the rolling-logit model, compared with the benchmark model, exhibited higher overall accuracy. The successful predictive performances were attributed to a recall mechanism in the rolling-logit model, measuring a corporation's risks on the basis of consistent information across time. However, this study had limitations due to the selection of the sample and variables. These limitations suggest that future research could validate its applicability in other markets or select other explanatory variables in order to improve the predictive ability.
|