Change Font: A A A A Contact Us What's New FAQs Subscribe ADB.org home
HomePublicationsCatalogCoordinated Failure? A Cross-Country Bank Failure Prediction ModelEndnotes

Endnotes

1 Beaver (1966) and (1968) was a pioneer in this approach. Altman (1968) proposed an alternate approach, multiple discriminant analysis (MDA), but later studies, such as Martin (1977) concluded that the distribution assumptions of the logit model are more reasonable.

2 Meyer and Pifer (1970) found that financial variables are good predictors of failure even when the cause of failure is embezzlement or other financial irregularities. Recent work by Hillegeist, Keating, Cram and Lundstedt (2003) recommends including market based measures in addition to financial variables, but this is not feasible here as most Indonesian banks do not have real market data available for analysis: there are only 26 banks listed on the Indonesian Stock Market and of these only 8 are actively traded.

3 Consolidation in the Japanese banking sector since 2001 means that it is not possible to construct a consistent sample of banks extending beyond 2001.

4 The logistic function looks very much like the cumulative normal function but is much easier to calculate as is does not require evaluating an integral.

5 See Santoso (2004), Rencher (1995) and Qurriyani (2000) for a detailed description of this method.

6 If the probability of survival, or non-bankruptcy is , then the ratio of the probability of bankruptcy to survival, , is the odds ratio. The log of the odds ratio is not only linear in X, but also (from the estimation view point) linear in the parameters. Thus the relationship between the logistic coefficients and the odds ratios can be expressed as odds ratio = exp(logistic coefficients).

7 Note that although the odds ratio may be negligible, it can still be statistically significant, as the coefficient estimates from which the odds ratios were derived was statistically significantly different from 0

8 Since the number of covariate patterns in our data set is equal to the number of observations, the Hosmer and Lemeshow test is preferred to the Pearson chi-square goodness of fit test.

Download this Discussion Paper [ PDF 286.6KB| 22 pages ].




[previous chapter]


Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [0] comment(s) for this entry. Post a comment.

    The views expressed in this paper are the views of the authors and do not necessarily reflect the views or policies of the Asian Development Bank Institute (ADBI), the Asian Development Bank (ADB), its Board of Directors, or the governments they represent. ADBI does not guarantee the accuracy of the data included in this paper and accepts no responsibility for any consequences of their use. Terminology used may not necessarily be consistent with ADB official terms.

    Back to Top 
    © 2012 Asian Development Bank Institute.