Modeling the dynamic patterns of banking and non-banking financial intermediaries’ performance

Bukhtiarova, A. H., Semenoh, A. Y., Mordan, Y. Y., Kremen, V. M., & Balatskyi, Y. O. (2022).

This article aims at quantifying the financial stability and resilience to illegal practices by financial intermediaries like banks, insurance companies and credit unions. Viscovery SOMine is used to cluster indicator variables obtained using the Harrington desirability function approach applied to collections of values describing different subfields.

The trajectories of companies’ financial architecture in the real economy

Shkolnyk, I. O., Mentel, U., Bukhtiarova, A. H., & Dushak, M. (2020).

In order to analyze the development of the 200 largest Ukrainian companies, their financial data for the years 2007 to 2017 are clustered with Viscovery SOMine. The clusters are ranked from best to worst according to financial indicators. The migration of 22 companies of interest between the clusters is analyzed.

Estimation of the production potential of Ukraine's regions using Kohonen neural network

Liashenko, O., Kravets, T., & Verhai, T. (2018).

Twenty-four Ukrainian regions are analyzed with respect to investment volumes and agricultural and industrial production. Viscovery SOMine is used to identify the most competitive regions and analyze similarities and dissimilarities to less competitive ones.

A comparative analysis of Serbia and the EU member states in the context of networked readiness index values

Soldić-Aleksić, J., & Stankić, R. (2015). Economic Annals, 60(206), 45-86.

This study analyzes the acceptance rate, level of use and economic impact of information technologies in EU member states and in Serbia and subsumes this information in a networked readiness index. The cluster model obtained using Viscovery SOMine shows that Serbia is in the group with the smallest networked readiness, together with several other South and East European countries.

Exploiting the self-organizing financial stability map

Sarlin, P. (2013). Engineering Applications of Artificial Intelligence, 26(5-6), 1532-1539.

This article enhances the information-extraction capabilities of the self-organizing financial stability map, a tool to monitor the financial stability of economies. With Viscovery SOMine, a probabilistic modeling of state transitions, contagion analysis and outlier detection are implemented.

Chance discovery with self-organizing maps: discovering imbalances in financial networks

Sarlin, P. (2013). In Advances in Chance Discovery (pp. 49-61). Springer, Berlin, Heidelberg.

This article discusses possibilities to use self-organizing maps for chance discovery. The Viscovery model first introduced by Sarlin et. al. in "Mapping the state of financial stability" (2011) is combined with a network analysis to reveal bilateral exposures and possible shock propagations between the financial systems of nations.

A framework for state transitions on the self-organizing map: some temporal financial applications

Sarlin, P., Yao, Z., & Eklund, T. (2012). Intelligent Systems in Accounting, Finance and Management, 19(3), 189-203.

A method for analyzing temporal transitions on self-organizing maps is introduced. Viscovery SOMine is used to apply this method to 1) the development of financial performance of banks and 2) the time-varying hazard of currency crises in emerging markets.

Visualizing indicators of debt crisis in a lower dimension: a self-organizing maps approach

Sarlin, P. (2012). In Handbook of Research on Computational Science and Engineering: Theory and Practice (pp. 414-431). IGI Global.

This article explores how nations' debt crises can be analyzed with self-organizing maps. Viscovery SOMine is used to create a model on annual debt data for 42 emerging economies between 1981 and 2004, which can be used to identify high risk situations.

Clustering the changing nature of currency crises in emerging markets: an exploration with self-organising maps

Sarlin, P. (2011). International Journal of Computational Economics and Econometrics, 2(1), 24-46.

The changing nature of currency crises in emerging market economies is analyzed. The Viscovery SOMine cluster model shows that crises in the late 1990s have a different character than those in the 1970s and 1980s and showed weaker warning signals.

Evaluating a self-organizing map for clustering and visualizing optimum currency area criteria

Sarlin, P. (2011). Economics Bulletin, 31(2), 1483-1495.

This analysis searches for the economically optimal combination of countries to have a single currency. To this end, the countries of the European monetary union, other EU members and a control group (Japan, Canada, Turkey and Norway) are clustered with Viscovery SOMine with respect to various currency indicators, such as interest rates, inflation and OCA index. The EU countries Denmark, Sweden and Poland, as well as the non-EU country Norway, show fitting characteristics to join the monetary union.

Mapping the state of financial stability

Sarlin, P., & Peltonen, T. A. (2013). Journal of International Financial Markets, Institutions and Money, 26, 46-76.

The paper uses self-organizing maps (SOMs) for mapping the state of financial stability and visualizing the sources of systemic risks as well as for predicting systemic financial crises. The study uses Viscovery SOMine to train the SOM models and for the second-level clustering of the SOMs.

Fuzzy clustering of the self-organizing map: some applications on financial time series

Sarlin, P., & Eklund, T. (2011, June). In International Workshop on Self-Organizing Maps (pp. 40-50). Springer, Berlin, Heidelberg.

A fuzzy clustering algorithm is applied to a self-organizing map model to account for uncertainty in economic applications. Viscovery SOMine is used to create a model of currency crises, to determine crisp clusters as starting points for the fuzzy clustering algorithm, and to visualize the resulting fuzzy clusters.

Visual predictions of currency crises: a comparison of self-organizing maps with probit models

Sarlin, P., & Marghescu, D. (2010).

The aim of this article is to provide a better way to predict the occurrence of currency crises. To this end, Viscovery SOMine is used to build a classification model that provides better accuracy than probit classification, which is used as a benchmark. In addition, the self-organizing-map model is used for further explorative analysis, including the evolution of a countries’ vulnerability to develop a currency crisis over time.

Welfare states and social sustainability. An application of SEM and SOM in a virtuous circle environment

Hagfors, R., & Kajanoja, J. (2010).

Different approaches to social welfare and their connection with civic-mindedness are the focus of this paper. Viscovery SOMine is used to compare the welfare regimes of different countries and provide clusters of similar regimes. It is observed that the North European countries form a tight cluster with a well-functioning virtuous circle, which can be used as a model region for social sustainability.

A Visualization and clustering approach to analyzing the early warning signals of currency crises

Liu, S., Eklund, T., Collan, M., & Sarlin, P. (2010). In Business Intelligence in Economic Forecasting: Technologies and Techniques (pp. 65-81). IGI Global.

This article aims to isolate early warning signals for currency crises. Viscovery SOMine is used to cluster quarterly economic data for Finland between 1984 and 1994, to identify pre-crisis periods, and analyze risk factors.

SOM-based data analysis of speculative attacks' real effects

Arciniegas Rueda, I. E., & Arciniegas, F. A. (2009). Intelligent Data Analysis, 13(2), 261-300.

This paper searches for meaningful associations between economic, sociopolitical and legal background and the effects of speculative attacks. For this purpose, Viscovery SOMine is used for an exploratory analysis, which shows strong influences from the banking sector, regulatory framework and interest rates on the real effects of speculative attacks.

An East Asian community? — Regional and global dynamics: what do the numbers say?

Shanmuganathan, S. (2007).

This article attempts to quantify the potential for an East Asian economic union that would be similar to the European Union. Viscovery SOMine is used to inspect data from the World Bank and the ASEAN Financial and Macroeconomic Surveillance Unit and to compare the economic dispersity and development of East Asian countries compared to the countries in the European Union.

Self-organizing patterns in world poverty using multiple indicators of poverty, repression and corruption

Deboeck G. (2000). Neural Network World, International Journal on non-standard Computing and Artificial Intelligence 1-2/2000 [19], 239-254.

This paper maps world poverty based on multi-dimensions of poverty. These global maps are based on a well-established neural network algorithm implemented in the software tool Viscovery SOMine. They show world poverty based on similarity and dissimilarity in poverty structures.