Early warning systems and systemic banking crises in low income countries: A multinomial logit approach

Caggiano, Giovanni; Calice, Pietro; Leonida, Leone

doi:10.1016/j.jbankfin.2014.07.002

Cited by 98 publications

(84 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this paper we build on Caggiano et al (2014), who show that the above results hold for systemic banking crises as well for a sample of low income countries (LICs), and provide, to the best of our knowledge, the first systematic analysis of the role played by the duration of a systemic banking crisis in affecting the relative ability of multinomial and binomial logit models in correctly predicting the arrival of the crisis itself.…”

mentioning

confidence: 85%

“…Put differently, since banking crises are typically longlasting, in the periods following the onset of the crisis the economy is likely to be still in a state of crisis, and hence relevant economic variables behave differently from both "equilibrium" periods and the outbreak of a crisis. We call crisis duration bias this phenomenon related to the existence of a state of prolonged distress in the context of banking crises: not accounting for the existence of a third state in the economy, i.e., a period of adjustment after the outbreak of a banking crisis before going back to normal, might reduce the predictive power the estimated EWS (see Caggiano et al, 2014, for an analysis of the crisis duration bias in a sample of LICs).…”

Section: The Crisis Duration Biasmentioning

confidence: 99%

“…In building the EWS for predicting systemic banking crises, we consider the multinomial logit model, previously employed by Bussiere and Fratzscher (2006) in the context of currency crises and by Caggiano et al (2014) in the context of banking crises in LICs, as an alternative to the commonly used binomial models previously discussed. The estimated model returns a predicted measure of fragility of the banking sector, i.e.…”

Section: The Multinomial Logit Modelmentioning

confidence: 99%

“…The second step in our empirical strategy is to estimate the binomial logit models where the observations related to crisis years other than the first are (i) treated as non-crisis observations (Table 3) and (ii) discarded from the sample (Table 4) AUROC is then a measure of the signalling quality of the estimated EWS, which overcomes the problem of assuming a specific utility function for the policy maker in order to properly weight the costs associated to a given signal (see Hsieh andTurnbull,1996, andPeterson, 2013, for a general discussion of the AUROC; Juselius, 2013, andCaggiano et al, 2014, for an application to banking crises). A value of the AUROC equal to 0.5 refers 8 In order to capture the impact of the long run trend of money upon the probability of experiencing a banking crisis, we have expanded the set of regressors to include a measure of excess money.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Comparing logit-based early warning systems: Does the duration of systemic banking crises matter?

Caggiano

Calice

Leonida

et al. 2016

Journal of Empirical Finance

Self Cite

View full text Add to dashboard Cite

Highlights1. This paper examines systemic banking crises.2. We focus on the crisis duration bias 3. We compare multinomial and binomial logit models in correctly predicting crises.4. We consider a large and heterogeneous dataset. We find the multinomial logit model to outperform binomial models AbstractThis paper compares the performance of binomial and multinomial logit models in the context of building early warning systems (EWS) for systemic banking crises. We test the hypothesis that the predictive performance of binomial logit models is hampered by what we define as the crisis duration bias, arising from the decision to either treat crisis years after the onset of a crisis as non-crisis years or remove them altogether from the sample. In line with our hypothesis, results from a large sample of world economies suggest that i) the multinomial logit outperforms the binomial logit model in predicting systemic banking crises, and ii) the longer the average duration of the crisis in the sample, the larger the improvement.

show abstract

mentioning

confidence: 85%

Section: The Crisis Duration Biasmentioning

confidence: 99%

Section: The Multinomial Logit Modelmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Comparing logit-based early warning systems: Does the duration of systemic banking crises matter?

Caggiano

Calice

Leonida

et al. 2016

Journal of Empirical Finance

Self Cite

View full text Add to dashboard Cite

show abstract

“…Applications can be found in, for example, transportation (Yang et al, 2013), banking (Caggiano et al, 2014), fast-moving consumer goods and spending (Guadagn and Little, 1983), aviation (Wen et al, 2014), the auto industry (Cecchini, 2001) and the health and pharmaceutical industry (Tarlov and Ware, 1989). In this paper, a multinomial logit model (MNL) by McFadden (1981) is applied to investigate ship owners' port selection decisions in the BBR.…”

Section: Quantitative Multinomial Logit Regressionmentioning

confidence: 99%

Port choice strategies for container carriers in China: a case study of the Bohai Bay Rim port cluster

Yang

Wang

2016

IJSTL

View full text Add to dashboard Cite

Abstract:The ports of Dalian, Qingdao and Tianjin are the largest ports in the Bohai Bay Rim and as such are among the fastest growing container ports in China. This setting provides an opportunity to analyse the dynamics of competition involved in inter-port clusters. Competition in a port cluster can be examined from the perspective of port choice. To understand the key determinants of carriers' port choice strategies, on-site interviews were conducted in the fall of 2013. Having identified the factors involved, we performed multinomial logit regressions on them. The results indicated that foreign trade routes and destinations, improved design capacity, demand from the hinterland and whether or not there is a rapid boutique line, can sway a carrier's decision towards either a single port strategy or a combined strategy of making two or three stops at the ports of Dalian, Qingdao and Tianjin. As a result of this research, using combined qualitative and quantitative methodologies, a better understanding of the competition, development and governance of not only ports in a regional area, but throughout China itself, has been achieved. This has contributed overall to a better understanding of the dynamics and interrelationships involved between ports in the Bohai Bay Rim.

show abstract

Determinants of financial crises—An early warning system based on panel logit regression

Jemović

Marinković

2019

Int. J Fin Econ

View full text Add to dashboard Cite

Despite the fact that different types of financial crises are rooted in similar weaknesses of economy or may have common determinants, the very transmission mechanism may determine one category as leading or lagging behind others. We are focused on financial crises that necessarily have the features of systemic banking crises and assess econometric early warning system of 64 systemic banking crises that occurred in the period from 1977 to 2013. The paper employs two different procedures, based on panel logit regression. The dynamic discrete‐choice (binary) early warning model clearly outperformed the static model. The set of significant explanatory variables changed relative to the findings of the static model. The most significant predictor of the crises in the better performing model is deposit insurance system, followed by international reserves, M2‐to‐international reserves ratio, M2 multiplier, bank deposits, and bank reserves ratio. The statistical significance of the lagged variable confirmed the necessity to take the effect of crisis persistence into account.

show abstract

Early warning systems and systemic banking crises in low income countries: A multinomial logit approach

Cited by 98 publications

References 28 publications

Comparing logit-based early warning systems: Does the duration of systemic banking crises matter?

Comparing logit-based early warning systems: Does the duration of systemic banking crises matter?

Port choice strategies for container carriers in China: a case study of the Bohai Bay Rim port cluster

Determinants of financial crises—An early warning system based on panel logit regression

Contact Info

Product

Resources

About