Developing an early warning system to predict currency crises

Sevim, Cüneyt; Oztekin, Asil; Bali, Özkan; Gümüş, Serkan; Güreşen, Erkam

doi:10.1016/j.ejor.2014.02.047

Cited by 128 publications

(104 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Similarly, Sevim, Oztekim, Bali, Gumus, & Gursen (2014) develop an early warning system to predict currency crisis. These authors used three methods: logit regression, decision tree and artificial neural networks.…”

Section: The Methods Of Artificial Intelligencementioning

confidence: 99%

Financial Leading Indicators of Banking Distress: A Micro Prudential Approach - Evidence from Europe

Messai¹,

Gallali²

2015

ASS

View full text Add to dashboard Cite

Sovereign and Subprime crises have corrosive effects on European banking system. In this study, we aim to explain and predict the state of distress for 618 European banks for a five year period (2007)(2008)(2009)(2010)(2011). For this purpose, we applied early warning systems using traditional and developed methods: discriminant analysis, logistic regression and the artificial intelligence. Those methods aim to predict bank distress up to a year (two years) before it actually happened. Our study seeks to compare between these three methods and to choose the most appropriate for prediction. The key finding of this study demonstrates that the neural network method outperforms the other models. This result is too much useful for banks and help policy makers to do a better job in terms of regulatory reforms.

show abstract

Section: The Methods Of Artificial Intelligencementioning

confidence: 99%

Financial Leading Indicators of Banking Distress: A Micro Prudential Approach - Evidence from Europe

Messai¹,

Gallali²

2015

ASS

View full text Add to dashboard Cite

show abstract

“…While in case of the problem of crisis prediction, a class variable is usually present, unsupervised learning methods can still offer an important tool, with the most frequently used methods belonging to this group include selforganizing maps [23] and c-means clustering [12] In a recent and literature review, a complete picture of the application of machine learning in financial crisis prediction was given by Lin et al [11]. Additionally to the above mentioned two groups, they found statistics-based learning methods, for example logistic regression [24] or discriminant analysis [18], and other methods, such as genetic algorithms [19], as widely used in the crisis prediction literature.…”

Section: Machine Learning In Financial Risk Analysismentioning

confidence: 99%

“…A financial crisis is a state in a financial system that causes economic, social, and political costs [24] with disastrous effects on the affected economies. For this reason, developing various methods for the purpose of crisis prediction has been an important research focus in recent years.…”

Section: Introductionmentioning

confidence: 99%

Possibilistic Clustering for Crisis Prediction: Systemic Risk States and Membership Degrees

Mezei

Sarlin

2017

Proceedings of the 50th Hawaii International Conference on System Sciences (2017)

View full text Add to dashboard Cite

Research on understanding and predicting systemic financial risk has been of increasing importance in the recent years. A common approach is to build predictive models based on macro-financial vulnerability indicators to identify systemic risk at an early stage. In this article, we outline an approach for identifying different systemic risk states through possibilistic fuzzy clustering. Instead of directly using a supervised classification method, we aim at identifying coherent groups of vulnerability with macrofinancial indicators for pre-crisis data, and determine the level of risk for a new observation based on its similarity to the identified groups. The approach allows for differentiating among different possible pre-crisis states, and using this information for estimating the possibility of systemic risk. In this work, we compare different fuzzy clustering methods, as well as conduct an empirical exercise for European systemic banking crises.

show abstract

“…First, we benchmark several algorithms (i.e., logistic regression, neural network, rotation forest, random forest, stochastic adaboost, and kernel factory) to determine which algorithms work best on this problem. Second, we use information fusion to build a fusion model and determine which variables are important (Sevim et al 2014). Information fusion is a technique that intelligently combines the results of different algorithms (Oztekin et al 2013).…”

Section: Introductionmentioning

confidence: 99%

Identifying Soccer Players on Facebook Through Predictive Analytics

Bogaert¹,

Ballings²,

Hosten³

et al. 2017

Decision Analysis

View full text Add to dashboard Cite

Abstract. This study assesses the feasibility of identifying self-reported sports practitioners (soccer players) on Facebook. The main goal is to develop a system to support marketers with the decision as to which prospects to target for advertising purposes. To do so, we benchmark several algorithms (i.e., random forest, logistic regression, adaboost, rotation forest, neural networks, and kernel factory) using five times twofold cross-validation. To evaluate performance and variable importances, we build a fusion model, which combines the results of the other algorithms using the weighted average. This technique is also referred to as information-fusion sensitivity analysis. The results reveal that Facebook data provide a viable basis to come up with sports predictions as the predictive performance ranges from 72.01% to 80.43% for area under the receiver operating characteristic curve (AUC), from 81.96% to 83.95% for accuracy, and from 2.41 to 3.06 for top-decile lift. Our benchmark study indicates that stochastic adaboost, the fusion model, random forest, rotation forest, and regularized logistic regression are the best-performing algorithms. Furthermore, the results show that the most important variables are the average number of friends that play soccer, membership of a soccer group, and the number of favorite teams. We also assess the impact of our results on profitability by conducting a thorough sensitivity analysis. Our analysis reveals that our approach can be beneficial for a wide range of companies. The analysis and results in this study will assist sports brands with decisions regarding their implementation of targeted marketing approaches.

show abstract

Developing an early warning system to predict currency crises

Cited by 128 publications

References 41 publications

Financial Leading Indicators of Banking Distress: A Micro Prudential Approach - Evidence from Europe

Financial Leading Indicators of Banking Distress: A Micro Prudential Approach - Evidence from Europe

Possibilistic Clustering for Crisis Prediction: Systemic Risk States and Membership Degrees

Identifying Soccer Players on Facebook Through Predictive Analytics

Contact Info

Product

Resources

About