The use of fuzzy clustering algorithm and self-organizing neural networks for identifying potentially failing banks: an experimental study

Alam, Pervaiz; Booth, David E.; Lee, K.; Þórðarson, Þorvaldur

doi:10.1016/s0957-4174(99)00061-5

Cited by 107 publications

(56 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…See for instance Alam et al (2000), Arena (2008), Avkiran and Cai (2012), Curry et al (2003), Cole and Gunther (1995b), Cole and White (2012), Henebry (1997), Mayes (2005, 2009), Poghosyan andČihák (2009), Tatom andHouston (2011) and Wheelock and Wilson (2000). 27 Cole and White (2012), Curry et al (2003) and Wheelock and Wilson (2000) utilize such liquid assets.…”

mentioning

confidence: 99%

See 1 more Smart Citation

The Effectiveness of Capital Adequacy Measures in Predicting Bank Distress

Mayes

Stremmel

2012

SSRN Journal

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Terms of use: Documents in AbstractOur concern in this paper is two-fold: first to see whether the determinants of bank distress and failure have been any different in the GFC from previous years: second to see whether simple measures of capital adequacy outperform their risk-weighted counterparts as predictors, despite the focus on the later in the Basel framework. This paper examines bank distress within a large quarterly data set of FDIC-insured US banks from 1992 to 2012. We contrast the effects of risk-weighted and non-risk-weighted capital measures for various banking types using two estimation methods (logit and discrete survival time analysis). We predict bank failures and draw inferences about the stability of contributing bank characteristics. Our models incorporate CAMELS indicators that consider the bank-specific variables and macroeconomic conditions. We find that the non-risk-weighted capital measure, the adjusted leverage ratio, explains bank distress and failures best with considerable accuracy. Further, we find that the influence of the characteristics in the two methods differs only slightly. Also the characteristics of banks getting into bank distress are alike over time. That means that the familiar banking characteristics for identifying a distress-prone bank identified fragile banks effectively during the global crisis without new information and are likely to continue to work well in the future. Further, our findings suggest that the more complex a bank is the more effective is the leverage ratio compared to the risk-based capital ratio 2 .

show abstract

mentioning

confidence: 99%

“…For instance Henebry (1997), Mayes (2005, 2009) This idea is pursued by Alam et al (2000), Tatom andHouston (2011) andWhalen (1991).…”

mentioning

confidence: 99%

The Effectiveness of Capital Adequacy Measures in Predicting Bank Distress

Mayes

Stremmel

2012

SSRN Journal

View full text Add to dashboard Cite

show abstract

“…Kumar and Haynes (2003) explored firms' financial performance data in relation to the credit rating of a debt issue and found out that ANN is superior to the discriminant analysis model as it allows increasing the speed and efficiency of the rating process. In accordance with the results of Alam et al (2000), Wu et al (2006) and Kumar and Haynes (2003), ANNs are better suited to the analysis of a Banking sector, because they are flexible, but it can be too complex and slow.…”

Section: Comparison Of Techniquesmentioning

confidence: 53%

“…Alam et al (2000) compared the results derived by the closest hard partitioning of fuzzy clustering and by self-organizing neural networks. As an outcome a specific rating of relative bankruptcy likelihood was prepared.…”

Section: Comparison Of Techniquesmentioning

confidence: 99%

Evaluating banks performance using key financial indicators – a quantitative modeling of Russian banks

Sharma¹,

Shebalkov²,

Yukhanaev³

2016

jda

View full text Add to dashboard Cite

Since the financial crisis of 2008, risk based performance management has been one of the important indicator to determine the financial health of banks and financial institutions. This study relates to the problem within the Russian Banking sector for regulators to determine and reduce risks at the marco-level and assessing performance of banks at the micro-level. The objectives are: to analyse a range of performance indicators and to structure the Russian banking sector. To explore the structure of Russian Banking sector in terms of performance over the period 2000-2010, we took a sample of 1279 banks and the financial data which was in the HTML format was extracted through PHP programming. With the help of trend analysis, the period 2000-2010 was divided into four sub periods: the period of stabilization (2002-2004), substantial development (2004-2007), financial crisis (2007-2009) and moderate development (2009)(2010). Multivariate analysis were applied to classify the sample banks in these sub periods which provides evidence that despite the changes in the stage of development of the economy, the Russian Banking sector can be described with quantitative modeling. Naturally, the structural changes are affected by the described economic cycles, but these changes do not affect the determination capabilities of the model. In the period 2002-2004, nine types of banks are found. There are some prosperous as well as weak banks. During the period 2004-2007, banks had a chance to increase their profits; the banking sector became more differentiated -12 clusters are singled out. There is no doubt that the financial crisis also affected the banking industry; there were still 12 clusters in 2007-2009, but the majority were concentrated into a single cluster with low performance indicators. Finally, the Russian banking sector started its development in the period 2009-2010, uniting some bank clusters, 10 groups are found. The results indicated that through mathematical modelling, Russian banks could be rated as "rating groups" based on their performance which might be of particular interest to bank's managers, investors, credit analysts and bank regulators. Moreover, it could be emphasized that the changes in structure are not significant, as certain groups of banks can be found at any period of time. These groups or clusters can be referred to certain "rating groups" (from the banks with the best results to those with low results) and compared to international ratings. JEL Classifications: G17, G21, G32

show abstract

“…Here every object will be a part of every cluster. The membership weight that goes between 0:if it utterly doesn't belong to cluster and 1:if it utterly belongs to the cluster [9].…”

Section: Fuzzy Clusteringmentioning

confidence: 99%