The publication list below consists of: A. Articles in peer-reviewed scientific journals (12 pubs) B. Contributions to books & monographs (4) C. Articles in peer-reviewed conference proceeding (29) D. Working papers & other publications (6) E. Selected presentations & unpublished work in progress (6) A. Articles in peer-reviewed scientific journals Georgescu, I., Kinnunen, J., 2019, How the investor's risk preferences influence the optimal allocation in a credibilistic portfolio problem, Submitted. Click here for ArXiv version. A classical portfolio theory deals with finding the optimal proportion in
which an agent invests a wealth in a risk-free asset and a probabilistic risky
asset. Formulating and solving the problem depend on how the risk is
represented and how, combined with the utility function defines a notion of
expected utility. In this paper the risk is a fuzzy variable and the notion of
expected utility is defined in the setting of Liu's credibility theory. Thus
the portfolio choice problem is formulated as an optimization problem in which
the objective function is a credibilistic expected utility. Different
approximation calculation formulas for the optimal allocation of the
credibilistic risky asset are proved. These formulas contain two types of
parameters: various credibilistic moments associated with fuzzy variables
(expected value, variance, skewness and kurtosis) and the risk aversion,
prudence and temperance indicators of the utility function. Georgescu, I., Kinnunen, J., 2017, Optimal Prevention with Possibilistic Background Risk, International Journal of Management and Applied Science, ISSN: 2394-7926, Vol. 3, No. 4, April 2017, pp. 25-29. Possibility risk theory starts from the hypothesis that this is described by a fuzzy number and not by a random
variable, as in the traditional probabilistic modeling. This paper is an attempt of possibilistic approach to risk management
problem. In particular, the models proposed in this paper can be connected by the way the possibilistic risk influences the
probability of a loss, the dimensions of the loss or both of them simultaneously considered (i.e. self-insurance, self-protection,
or self-insurance-cum-protection). In the paper, there are studied two-period models of risk management in which the
possibilistic risk may appear in the first period or in the second period. The main results of the paper establish necessary and
sufficient conditions, or only sufficient, for the optimal management level effort to raise or to decrease as a result of adding
such a possibilistic risk.
Georgescu, I., Kinnunen, J., 2016, Mixed Models for Risk Aversion, Optimal Saving, and Prudence, Fuzzy Economic Review, The International Association for Fuzzy-Set Management and Economy, Vol. 2, No. 2, Nov 2016. The models of this paper refer to mixed risk situations: one parameter is a fuzzy number, and the other is a random variable. Three notions of mixed expected utility are proposed as a mathematical basis of these models. The results of the paper describe risk aversion and prudence of an agent in front of a risk situation with mixed parameters, and the changes of optimal saving as an effect of mixed risk. Georgescu, I., Kinnunen, J., 2016, Credibilistic Risk Aversion and Prudence, International Journal of Business Innovation and Research, Inderscience Publishers, Vol. 11, No. 1, pp. 146-160. Risk aversion and prudence are well-studied topics in probabilistic
risk theory. This paper uses credibility theory of B. Liu and Y. Liu to approach
these closely related concepts. The risk situations are modeled by fuzzy variables
and the indicators of risk aversion and prudence are defined in the context of
credibilistic expected utility theory. Calculation formulas for these indicators are
established and their properties are examined. The paper establishes relationships
between credibilistic risk aversion and prudence, as well as, between credibilistic
prudence and optimal precautionary saving. Georgescu, I., Kinnunen, J., 2015, Distances of Fuzzy Choice Functions, Journal of New Mathematics and Natural Computation, World Scientific Vol. 11, No. 3, 249 (2015). DOI: 10.1142/S1793005715500088 In
this paper we introduce four distances on the set of fuzzy choice
functions defined on a finite choice space. They are studied along with
four distances on the set of alternatives. The two types of distance
allow to investigate the way the changes in fuzzy preferences are
reflected in the changes of fuzzy choice associated with them. Also the
way the changes in fuzzy choices manifest themselves in changes in fuzzy
preferences are studied. The coefficient of normality of a fuzzy choice
function is defined as a measure of normality and its variation is
evaluated with respect to the variation of fuzzy choices. Finally, the
variation of some congruence indicators is evaluated as effect of the
changes in fuzzy choices. Georgescu, I., Kinnunen, J., 2013, A Risk Approach by Credibility Theory, Fuzzy Information and Engineering, Springer, Vol. 5, No. 4, pp. 399-416. DOI 10.1007/s12543-013-0154-0 This paper attempts to treat some topics of risk theory by means of credibility theory. We study the risk aversion of an agent faced with a situation of uncertainty represented by a discrete fuzzy variable, the relationship between stochastic dominance and credibilistic dominance, and an index of riskiness of discrete credibilistic gambles. In the framework of an optimal saving credibilistic model, the way the presence of risk modifies the level of optimal saving is studied. The main tool of our investigation is an operator defined by B. Liu and Y. K. Liu by which to a discrete fuzzy variable one associates a discrete random variable with the same expected value as the former.
Georgescu, I., Kinnunen, J., 2013, A New Notion of Possibilistic Covariance, Journal of New Mathematics and Natural Computation, World Scientific, Vol. 9, No.1, pp. 1-11. (On a top 10 most read articles of the journal as of January 2015) Possibilistic indicators of fuzzy numbers (expected value, variance, and covariance) are an efficient instrument in the modeling of uncertainty phenomena. Various models of uncertainty phenomena have led to several notions of variance and covariance. In particular, the possibilistic models of risk aversion previously studied by one of the authors imposed a notion of variance of a fuzzy number di.erent from those existing in the literature. In this paper a new notion of covariance of two fuzzy numbers corresponding to the possibilistic variance mentioned is studied. This possibilistic covariance can be used, e.g., in models of possibilistic risk aversion with many parameters. Georgescu, I., Kinnunen, J., 2012, A Mixed Portfolio Selection Problem, Advances in Intelligent and Soft Computing, 2012, Springer, Vol. 151, pp. 95-102. The mixed portfolio selection problem studied in this paper corresponds to a financial situation in which some return rates are mathematically described by random variables and others are described by fuzzy numbers. Both Markowitz probabilistic model and a possibilistic portfolio selection model are generalized. A calculation formula for the optimal solution of the portfolio problem and a formula which gives the minimum value of the associated risk are proved. Collan, M., Kinnunen, J., 2011, A Procedure for the Rapid Pre-acquisition Screening of Target Companies Using the Pay-off Method for Real Option Valuation, Journal of Real Options and Strategy, Vol. 4, No. 1, pp. 117-141.
Pre-acquisition target screening is a task that often needs to be made fast, but is still complex, especially due to the need to estimate the value of possible synergies arising from an acquisition. Acquisition synergies are highly uncertain, difficult to explicitly quantify, and require successful management actions to be realized. The valuation of synergies calls for methods that can handle both, high uncertainty and inexact information. This paper discusses how a rapid pre-acquisition screening of target companies can be performed with the pay-off method for real option valuation, treating the acquisition synergy as a real option available for the acquirer. The pay-off method is a simple, intuitive, and practitioner friendly method. We define acquisition synergy as the value arising from resource redeployments within the newly formed entity of the target and the acquirer, and as the value gained through the possible divesture of target company assets. The procedure presented can be used in the screening of prospective acquisition target companies. Georgescu, I., Kinnunen, J., 2011,
Multidimensional Possibilistic Risk Aversion, The International Journal of Mathematical and Computer Modelling, Elsevier, Vol. 54, No. 1-2, pp. 689–696.
This paper deals with the analysis of risk aversion of an agent
faced with a situation of uncertainty with several risk parameters. These risk
parameters are represented by fuzzy numbers and the attitude of the agent to
the risk situation by a multidimensional utility function. Risk aversion is
measured by the notion of generalized possibilistic risk premium. The main
result of the paper is an approximate calculation formula of generalized
possibilistic risk premium in terms of the utility function and of
possibilistic indicators (mean value and covariance).
Georgescu,
I., Kinnunen, J., 2011,
Credibility measures in portfolio analysis: From possibilistic to probabilistic models, Journal of Applied Operational Research, Tadbir Institute for Operational Research, Systems Design and Financial Services, Vol. 3, No. 2, pp. 91-102. This paper treats risk based on the notions of credibility measure
and credibility expected value. Firstly, the paper derives and discusses
the credibility expected value. Secondly, the paper presents a new
method of analysis of possibilistic portfolios. The new step is a
construction by which with a possibilistic portfolio one associates a
probabilistic portfolio. The problem solving of possibilistic portfolio
gets down to the problem solving of associated probabilistic portfolio. For the latter
one a variety of solving methods exist, from which we can choose the most appropriate one for the initial problem. Risk evaluation in the
context of probabilistic portfolio leads to an understanding of risk for
the possibilistic portfolio. The paper presents an illustrative case of
a venture capitalist firm, which needs to solve the shares of its
budget to be invested to start-up companies.
Collan, M., Kinnunen, J., 2009, Acquisition Strategy and Real Options, The IUP Journal of Business Strategy,
Vol. VI (3&4), September-December, pp. 45-65. Link to Working paper PDF This paper presents strategic level real options that acquiring
companies have in the corporate acquisitions process. The real options
presented are different from those residing within the acquisition candidate
companies as stand-alones. The paper further presents acquisition synergies as
real options and as sequential real options to acquisition timing, and the
rather special case of acquiring businesses with divestible non-core assets.
Models for the valuation of the aforementioned real options are discussed with
special focus on input variable values. The paper illustrates the non-core
business acquisition issue with a case of acquisition and partial divestment of
Partek by Kone. Georgescu, I., Kinnunen, J., Casademunt, A.M.L., 2015, Possibilistic Models of Risk Management, in Kahraman and S. Çevik Onar (eds.), Intelligent Techniques in Engineering Management, Intelligent Systems Reference Library 87, pp. 21-44, DOI: 10.1007/978-3-319-17906-3_2, Springer
Georgescu, I., Kinnunen, J., 2012, Mixed Multidimensional Risk Aversion, in R.-E. Precup, S. Kovacs, S. Preitl and E. M. Petriu (Eds.), Applied Computational Intelligence in Engineering and Information Technology - Topics in Intelligent Engineering and Informatics, 2012, Vol. 1, pp. 39-50, DOI: 10.1007/978-3-642-28305-5_3, Springer. The topic treated in this chapter is the risk aversion of an agent in front of a situation of uncertainty with many risk parameters. We will study a general model of risk aversion in which some parameters are probabilistically described (by random variables) and others are possibilistically described (by fuzzy numbers). For the construction of this model, firstly, mixed expected utility, a notion, which unifies probabilistic and possibilistic aspects of expected utility theory is introduced. The notion of mixed risk premium vector is introduced as a measure of risk aversion with mixed parameters. The main result of the chapter is an approximate calculation formula for mixed risk premium vector. Lastly, our model is applied in the evaluation of risk aversion in grid computing. AssessGrid, 2008, Advanced Risk Assessment, EU-Project AssessGrid Deliverable, September, 2008. (Co-author and co-editor). The deliverable describes the risk assessment approach and models developed in the EU-funded project, AssessGrid, aiming at developing software and possible standards to assess risks in grid computing. The deliverable presents the risk assessment from the view of the provider of grid computing services. The focus of the assessment is on the evaluation of business risk, the risks of violating service level agreements, SLAs. The deliverable presents state-of-the-art review of risk assessment in grid environment, presents a statistical (Bayesian) approach and approximate approaches (a hybrid model to include probabilistic and possibilistic elements, fuzzy Bayesian, and a linguistic approach) to be applied in dynamic and predictive risk assessment. Kinnunen, J., 2002, Industry Foresight [with Software Agents: an Energy Case], Masters Thesis in economics, University of Vaasa. C. Articles in peer-reviewed conference proceedings Kinnunen, J., Androniceanu, A., Georgescu, I., 2019, Impact of Economic and Political Systems on Human Development Index: A Computational Evaluation, Submitted. There are various complex interdependences
between economic and political factors and United Nation’s human development
index, HDI. By using Bertelsmann Foundation’s data on 123
countries-in-transition with respect to their level of market economy,
democracy and governance quality, the most important factors affecting HDI are
under scrutiny. The purpose is to find out, which characteristics of countries
predict or are related to the level of HDI as reflected in the atypical set of
quantitative and qualitative statistical data. The methods used to process the
data and analyse correlations are artificial neural networks and multiple
correspondence analysis. The original contributions of this research include:
proving the effectiveness of neural networks; in general, developed marked economy
and democracy are related to high levels of HDI; more specifically, in addition
to national income, GNI, and educational variables also underlying HDI, the
measures of equal opportunities and lack of socio-economic barriers, among
others, are found important predictors for high HDI levels. Kinnunen, J., Georgescu, I., Androniceanu, A., 2019, EU Digitalization: A Multivariate Approach, Submitted to the 10th Scandinavian Conference on Information Systems (SCIS2019), August 11-14, Nokia, Finland. [The paper does cluster-wise analysis on similar, but slightly different clusters as the earlier paper below, and further, it formulates principal components for the whole country set, as well as for all clusters] European Union is heading towards a digital single market based on its Digital Single Market strate-gy accepted in 2015. European countries have started their national digitalization strategies from very different levels and the gap is considerable between frontrunners and challengers. In this paper, multivariate data analysis techniques are used: 29 European countries are firstly grouped by clustering analysis using ten variables of different aspects of digitalization. Five clusters of similar countries are obtained. The clusters, as well as, the countries within each cluster show consistent performance roughly over all indicators. Secondly, principal component regression formulas are constructed based on principal component analysis. This is done for the EU as a whole and, separately, for each cluster. Androniceanu, A., Georgescu, I., Kinnunen, J., 2019, Digitalization Clusters within the European Union, Accepted to the 33rd IBIMA Conference, April 10-11, Granada, Spain. Digitization
is strategic investment for states, as well as, for enterprises, to
boost the development by increasing efficiency and productivity and
generating substantial savings. The aim of this research is to identify
the digitalization clusters and to analyze the factors and impacts of
digitalization within the European Union. Our research is based on 10
indicators from 29 European countries. We used the EUROSTAT database for
the year 2018. The selected indicators are processed using the k-means
algorithm to obtain five clusters. The results of the research show
significant differences between the EU member states and clusters. Based
on the selected indicators and key parameters, the authors characterize
digitalization in the EU countries and made an original processing
using the k-mean algorithm. This is the first approach of this type, if
we refer to the literature. According to the latest report published by
the European Commission, Romania and Bulgaria, which form one cluster,
show the fewest changes in the digital domain, while the Nordic
countries, which all belong to another cluster, are on the opposite end
of the spectrum. Kinnunen, J., 2018, Islamic Countries Clustered and Analysed by Ecomic and Political Freedoms and GDP, 6th International Islam and Liberty Conference: Building an Islamic Case for Open Markets, November 14-15, 2018, Islamabad, Pakistan. Georgescu, I., Kinnunen, J., 2018, The influence of prudence and temperance on the credibilistic portfolio optimization, Proceedings of the 18^{th} IEEE International Symposium on Computational Intelligence and Informatics, November 21-22, 2018, Budapest, Hungary, pp. 199-204. ISBN 978-1-7281-1116-2 In this paper is studied the question of finding the optimum proportion in which an agent invests a wealth in a riskfree asset and a credibilistic risky asset. The credibilistic portfolio problem is formulated using the notion of credibilistic expected utility. Different approximate calculation formulas for the optimal allocation of the credibilistic risky asset are formulated depending on the credibilistic moments and risk aversion, prudence and temperance. Georgescu, I., Androniceanu, A., Kinnunen, J., 2018, A Computational Analysis of Economic Freedom Indicators and GDP in EU States, 17th International Conference on Informatics in Economy, May 17-20, 2018, Lasi, Romania. This paper studies the relationship between 12 freedom indicators and GDP per capita of the 28 European Union countries as of 2016. OECD’s GDP data for the EU countries is analysed together with the freedom indicators behind the Index of Economic Freedom. The data is studied using hierarchical clustering method and principal component analysis. The interpretations of the obtained clusters and the main findings are presented and graphically analyzed. Kinnunen, J., Georgescu, I., Tamminen, L., 2017, Do economic freedoms create national wealth? Clustering and principal component analyses of freedom variables and GDP, Proceedings of The Real Options Workshop, October 3-October 4, 2017, Lappeenranta, Finland. This
paper studies the relationship of economic freedom variables, GDP per
capita, and GDP growth. The OECD’s data for GDP of 36 countries are
firstly matched with 10 freedom variables behind the Index of Economic
Freedom. The normalized data is then clustered by self-organizing maps
to get four clusters of countries. The constructed clusters are
intuitively explainable, but principal component analysis is used to
support the interpretation of each cluster. The paper discusses a real
options perspective, how economic freedoms can open up wealth-creating
options and increase GDP growth, and in which circumstances this may not
be the case. The purpose of the study is to lay a ground for later
exploration of links from economic freedoms to material and non-material
well-being of which at least the first one clearly is affected by GDP. Kinnunen, J., Georgescu, I., 2017, VC-Portfolio Selection under Different Risk Models, Accepted to the 21^{st }Annual International Conference on Real Options, June 31- July 1, 2017, Boston, MA, USA. This paper views a venture capitalist's (VC's) portfolio selection
problem under three types of uncertainties corresponding to
probabilistic, possibilistic, and credibilistic risks. A VC faces such a
complex situation, when several potential target companies are under
analysis and some are handled in probabilistic terms while others call
for a possibilistic or a credibilistic treatment. A possibilistic
instead of probabilistic treatment is suitable for cases, where
uncertainty is very high. This can be due to lack of available
statistical information, because investment targets are often privately
owned small companies with limited public information, possibly totally
without past sales, without market values, there don't exist comparable
firms to allow comparables-based valuations, and their value largely
depends on intangibles and their strategic future actions. Credibilistic
approach also has similar benefits over a probabilistic approach and
certain computation benefits over a possibilistic approach. This paper
unifies probabilistic, possibilistic ,and credibilistic portfolio
selection models resulting in the optimal solution of the 3-component
portfolio problem faced by a VC. The paper discusses the added value and
the usefulness of the the approach in VC framework and presents
simulation analysis and a calculation example to demonstrate the
practicality of the approach. Georgescu, I., Kinnunen, J., 2017, Optimal Prevention with Possibilistic Background Risk, Proceedings of The IRES International Conference on Science, Technology and Management, February 13-14, 2017, Marrakech, Morocco. ISBN: 978-93-86083-34-0. Possibility risk theory starts from the hypothesis that this is described by a fuzzy number and not by a random variable, as in the traditional probabilistic modeling. This paper is an attempt of possibilistic approach to risk management problem. In particular, the models proposed in this paper can be connected by the way the possibilistic risk influences the probability of a loss, the dimensions of the loss or both of them simultaneously considered (i.e. self-insurance, self-protection, or self-insurance-cum-protection). In the paper, there are studied two-period models of risk management in which the possibilistic risk may appear in the first period or in the second period. The main results of the paper establish necessary and sufficient conditions, or only sufficient, for the optimal management level effort to raise or to decrease as a result of adding such a possibilistic risk. Georgescu, I., Kinnunen, J., 2016, Mixed Models for Risk Aversion, Proceedings of the North-European Society for Adaptive and Intelligent Systems' NSAIS'16 Workshop on Adaptive and Intelligent Systems, August 16-19, 2016, Lappeenranta, Finland. The models of this paper refer to mixed risk
situations: one parameter is a fuzzy number, and the other is a random
variable. Three notions of mixed expected utility are proposed as a
mathematical basis of these models. The results of the paper describe risk
aversion and prudence of an agent in front of a risk situation with mixed
parameters, and the changes of optimal saving as an effect of mixed risk.
Kinnunen, J., 2016, Economic Freedom Measures and Global Wealth Distribution. Proceedings of the 10^{th }RSSIA Summer Seminar on New Institutional Analysis, July 2-6, 2016, Moscow, Russia. Global
inequality is inherently a hot topic of economic and political debates.
Attitudes towards wealth and income inequalities depend radically on
the point of view: some think that some intervention by a state or
supra-national actors is required; others think that regulation and lack
of free trade have caused the inequalities, while the inequalities are
rather seen as incentives. This study focuses on the relationship
between global wealth inequalities and economic freedom variables using
clustering analysis. The starting point is Credit Suisse’s global wealth
data and the ten variables behind the Wall Street Journal’s and
Heritage Foundation’s Index of Economic Freedom: property rights,
freedom from corruption, fiscal freedom, government spending, business
freedom, labor freedom, monetary freedom, trade freedom, investment
freedom, and financial freedom. The freedom as well as the wealth data
are averaged over three years 2013-2015 to get better generalizability.
Firstly, the economic freedom indicators, which go hand in hand with the
country-specific wealth measures, are identified. Secondly, the
relationship of the economic freedom and wealth distribution is under
study.
Georgescu, I., Kinnunen, J., 2016, New Mixed Models of Optimal Saving, Proceedings of the 11^{th} IEEE International Symposium on Applied Computational Intelligence and Informatics, May 12-14, 2016, Timisoara, Romania, pp. 407-412. ISBN: 978-1-5090-2379-0 This paper studies mixed models of saving with three types of risks: labor income risk, interest rate risk, and background risk. Each risk can be probabilistically modeled by a random variable or possibilistically modeled by a fuzzy number. For each model a notion of precautionary saving is deﬁned in order to measure the effect that various types of risk have on optimal saving. The main results establish necessary and sufﬁcient conditions on consumer’s prudence when risk parameters are introduced. Georgescu, I., Kinnunen, J., 2015, Precautionary Saving with Possibilistic Background Risk, Proceedings of 16^{th} IEEE International Symposium on Computational Intelligence and Informatics, Budapest, Hungary, November 19-20, 2015. This paper studies a mixed model of saving with three risks: labor income and interest rate risks are random variables and background risk is a fuzzy number. Two notions of precautionary saving are introduced, measuring the effect of these types of risk on optimal saving and necessary and sufﬁcient conditions on their non–negativity are proved. This leads to two concepts of consumer’s prudence when new risk parameters are introduced. The main results of the paper characterize the two concepts of prudence by conditions of partial derivatives of consumer’s utility function. Kinnunen, J., Georgescu, I., 2015, Global Inequality, Economic Freedoms, and Real Options Thinking, presented in the 19^{th }Annual International Conference on Real Options, Athens, Greece, June 27-20, 2015. Global
inequality has surged recently as a political and economic issue of which
Piketty phenomenon has been just one example. Income or wealth inequality is
seen by some to require intervention from government and transnational bodies,
whereas others see government intervention and lack of free trade as the reason
for inequality, while finding some inequality good for incentive purposes. This
paper studies the relationship of economic freedoms, flexibility, and wealth
inequality. We start from Credit Suisse’s wealth data published yearly in Global Wealth
Report. We extend the typical approach focusing on net wealth distributions,
firstly, by presenting similar calculations using the value of total assets,
which includes both owner's equity and borrowed capital and, secondly, consider
the real options view on the levered agents’ total assets and argue for the benefits of
using total asset value instead of net wealth value and we suggest to use
real-options-based total value distributions for more meaningful interpretation
of inequality, global or national. In this approach, debt is not seen as a
negative cost, because it rather reflects opportunities offered by robust
financial markets and access to credit. Finally, we use clustering analysis to
study how the inequality of total assets is related to ten variables behind the
Heritage Index of Economic Freedom: property rights, freedom
from corruption, fiscal freedom, government spending, business freedom, labor
freedom, monetary freedom, trade freedom, investment freedom, and financial
freedom. This leads to new insights into global inequality and economic
opportunity and opens up several interesting new research questions. Kinnunen, J., Georgescu, I., 2014, Clustering Analysis of Real Option Value and Financial Ratios Using Self-Organizing Map, Accepted to the 18^{th }Annual International Conference on Real Options, Medellin, Colombia, July 23-26, 2045.
Georgescu, I., Kinnunen, J., 2013, A Credibilistic Approach to Risk Aversion and Prudence, Proceedings of the Finnish Operations Research Society 40^{th }Anniversary Workshop (FORS40) on Optimization and Decision-Making, Lappeenranta, Finland, August 20-21, 2013, pp. 72-77. Risk aversion and prudence are deeply studied topics in probabilistic risk theory. In this paper an approach of the two topics by credibility theory of B. Liu and Y. Liu is attempted. The risk situations are modeled by fuzzy variables and the indicators of risk aversion and prudence are defined in the context of credibilistic expected utility theory. Properties of these indicators are studied and approximate calculation formulas are established.
Georgescu, I., Kinnunen, J., 2012,
A Probabilistic Approach to Possibilistic Risk Aversion, Proceedings of the 7th IEEE International Conference on Computer Sciences and Convergence Information Technology (ICCIT 2012), Seoul, Korea, December 3-5, 2012, pp. 342-345. This paper proposes an approach to possibilistic risk aversion by means of probabilistic concepts. A notion of probabilistic risk premium is defined as the probabilistic risk premia associated with the uniform distributions on the level sets of a fuzzy number. We prove an approximate computation formula for this indicator and a Pratt-type theorem to compare two agents's possibilistic risk aversions.
Georgescu, I., Kinnunen, J., 2012, Possibilistic Risk Aversion and Its Indicators, Proceedings of the 11th WSEAS International Conference on Applied Computer and Applied Computational Science (ACASOS 2012), Rovaniemi, Finland, April 18-20, 2012, Recent Researches in Applied Computers & Computational Science, pp. 178-183. In the traditional treatment, risk situations are modeled by random variables. This paper focuses on risk situations described by fuzzy numbers. The goal of the paper is to define and characterize possibilistic risk aversion and study some of its indicators. Georgescu, I., Kinnunen, J., 2012, A Generalized 3-Component Portfolio Selection Model, Proceedings of the 11th WSEAS International Conference on Artificial Intelligence, Knowledge Engineering and Data Bases (AIKED 2012), Cambridge, UK, February 22-24, 2012, Recent Researches in Artificial Intelligence and Database Management, pp. 142-147. In this paper we study a portfolio selection problem corresponding to a financial situation characterized by three components: some returns are mathematically described by random variables, others by fuzzy numbers, and a third group of returns by discrete fuzzy variables. The proposed model unifies probabilistiv, possibilistic, and credibilistic aspects of portfolio selection. Both Markowitz probabilistic model and a possibilistic portfolio selection model are generalized. A calculation formula for the optimal solution of the portfolio problem and a formula, which gives the minimum value of the associated risk are proved. Georgescu, I., Kinnunen, J., 2012, A Mixed Portfolio Selection Problem, Proceedings of to the International IEEE Symposium on Distributed Computing and Artificial Intelligence (DCAI 2012), Salamanca, Spain, March 28-30, 2012. Advances in Intelligent and Soft Computing, 2012, Vol. 151, Distributed Computing and Artificial Intelligence, pp. 95-102, Springer. The mixed portfolio selection problem studied in this paper corresponds to a financial situation in which some return rates are mathematically described by random variables and others are described by fuzzy numbers. Both Markowitz probabilistic model and a possibilistic portfolio selection model are generalized. A calculation formula for the optimal solution of the portfolio problem and a formula which gives the minimum value of the associated risk are proved. Georgescu, I., Kinnunen, J., 2011, Credibilistic Index of Riskiness, Proceedings of the 12th IEEE International Symposium on Computational Intelligence and Informatics (CINTI 2011), Budapest, Hungary, November 21-22, 2011. This paper treats risk theory by means of credibility theory. The focus is on risk situations represented by discrete fuzzy variables. They can be useful to an agent in the decision-making process of credibilistic risk problems. An index of riskiness of Aumann-Serrano type for discrete credibilistic gambles is defined and the proofs are provided. Georgescu, I., Kinnunen, J., 2011, Modeling the Risk by Credibility Theory, Proceedings of the 3rd International IEEE Conference
on Advanced Management Science (ICAMS 2011), Kuala Lumpur, Malaysia,
November 4-6, 2011, International Proceedings of Economics Development and Research, Vol. 19, pp. 15-19. This
paper is an attempt to treat some topics of risk theory by means of credibility
theory. The risk aversion of an agent faced with a situation of uncertainty
represented by a discrete fuzzy variable and the relationship between stochastic
dominance and credibilistic dominance are under study. Georgescu, I., Kinnunen, J., 2011, Possibilistic Risk Aversion with Many Parameters, Proceedings of the International Conference on Computational Science (ICCS'11), Singapore, June 1-3, 2011; Procedia Computer Science, Elsevier, Vol. 4, pp. 1735-1744. In
this paper a possibilistic model of risk aversion with several
parameters is proposed. The notion of possibilistic risk premium vector
is introduced as a measure of an agent’s risk aversion to a situation
with several risk parameters. The main result of the paper is an
approximate calculation formula of this indicator. The way we can apply
this model in risk aversion evaluation in grid computing is sketched
out. Georgescu, I., Kinnunen, J., 2011,
Multidimensional Risk Aversion with Mixed Parameters, Proceedings of the 6^{th} IEEE International Symposium on Applied Computational Intelligence and Informatics (SACI 2011), Timisora, Romania, May 19-21, 2011. In this paper we propose an approach of risk aversion for the situations with many risk parameters. Some of the parameters are described probabilistically, and others possibilistically. We introduce mixed risk premium vector, a notion which combines probabilistic and possibilistic aspects of risk aversion. The main result of the paper is a formula for the calculation of the mixed risk premium vector. Our model can be applied for the evaluation of risk aversion in grid computing. Georgescu, I., Kinnunen, J., 2010, Multidimensional Possibilistic Risk Aversion, Proceedings of the 11^{th} International Symposium on Computational Intelligence and Informatics (CINTI'10), Budapest, Hungary, November 18-20, ISBN 978-1-4244-9278-7, IEEE Catalog Number CFP1024M-PRT, pp. 163-168. In this paper the notion of generalized possibilistic risk premium is introduced as a measure of the risk aversion of an agent faced with several components of possibilistic risk. The main result of the paper is a formula for the calculation of the generalized possibilistic risk premium expressed in terms of a utility function and of some possibilistic indicators. Georgescu,
I., Kinnunen, J., 2010,
Credibility Measures in Portfolio Analysis, Proceedings of the 2^{nd}
International Conference on Applied Operational Research (ICAOR'10), Turku,
Finland, August 25-27, Lecture Notes in Management Science, Vol. 2, pp. 6-18.
This paper treats risk based on the notions of credibility measure and credibility expected value. Firstly, the paper derives and discusses the credibility expected value. Secondly, the paper presents the definition and analysis of possibilistic portfolios. With a possibilistic portfolio a probabilistic portfolio is canonically associated. Risk evaluation in the context of probabilistic portfolio leads to an understanding of risk for the possibilistic portfolio. Kinnunen, J., 2010, Valuing M&A Synergies as (Fuzzy) Real Options, Proceedings (CD-ROM) of the 14^{th} Annual International Conference on Real Options, Rome, Italy, June 16-19. This paper views operating synergies as real options that acquiring companies have in the post-acquisition M&A process. The paper builds on the synergistic restructuring theory stating that both acquisitions and divestitures are wealth creating activities and takes a holistic view on acquisition synergies, which arise both from resource redeployments between the acquirer and the acquisition target company and the executed divestitures of target’s assets within the post-acquisition process. We present a procedure to ex-ante calculate a first approximation for the value of the synergies in the screening stage of the M&A process. We argue that synergies are highly uncertain and require significant management actions and, for that reason, an appropriate method for the valuation is the fuzzy pay-off method, which is presented as an integrated part of a decision support system built for the screening of potential acquisition targets. The paper also discusses the ordering of acquisition candidates according to their total value based on the presented fuzzy measure. Georgescu, I., Kinnunen, J., 2010, A New Notion of Possibilistic Covariance, Proceedings of the 6^{th} International Conference on Economic Cybernetic Analysis:
Global Crisis Effects and the Patterns of Economic Recovery (GCER 2011) Bucharest, Romania, May 20-21, pp. 159-167.
In this paper a new covariance of
fuzzy numbers is studied. It corresponds to a notion of possibilistic variance
used previously by the authors for the evaluation of possibilistic risk
premium. Kinnunen, J., Collan, M., 2009, Supporting the Screening of Corporate Acquisition Targets, Proceedings of the 42^{nd} International Conference on System Sciences (HICSS'09), Waikoloa, Hawaii, January 5-8. The paper discusses the screening of potential corporate acquisition targets from the acquiring company perspective. A structured process to screen acquisition targets using both, financial and non-financial data and a developed prototype DSS to support the screening process are presented. Collan, M., Kinnunen, J., 2008, Strategic Level Real Options in Corporate Acquisitions, Proceedings of the 1^{st} International Conference on Applied Operational
Research (ICAOR'08), Yerevan, Armenia, September 15-17.
In the paper we present strategic level real options that acquiring companies have in the corporate acquisitions process. The real options presented are such that exist on the strategic level and are different from the real options that reside within the acquisition candidate companies as stand-alone. We present acquisition synergies as real options and strategy level real options created in the acquisition as sequential real options to acquisition timing. Evaluation of target companies that includes the aforementioned real options is discussed. D. Working papers & other publications Georgescu, I., Kinnunen, J., 2011, Modeling the Risk by Credibility Theory, IAMSR Research Report 6/2011. This
paper
is an attempt to treat some topics of risk theory by means of credibility
theory. The risk aversion of an agent faced with a situation of uncertainty
represented by a discrete fuzzy variable, the relationship between stochastic
dominance and credibilistic dominance and an index of riskiness of discrete
credibilistic gambles is under study.
Georgescu, I., Kinnunen, J., 2011, Multidimensional Risk Aversion with Mixed Parameters, IAMSR Research Report 3/2011. In this paper we propose an approach of risk aversion for the situations with many risk parameters. Some of the parameters are described probabilistically, and others possibilistically. We introduce mixed risk premium vector, a notion which combines probabilistic and possibilistic aspects of risk aversion. The main result of the paper is a formula for the calculation of the mixed risk premium vector. Our model can be applied for the evaluation of risk aversion in grid computing. Georgescu, I., Kinnunen, J., 2010, Multidimensional Possibilistic Risk Aversion, IAMSR Research Report 3/2010. In this paper the notion of generalized possibilistic risk premium is introduced as a measure of the risk aversion of an agent faced with several components of possibilistic risk. The main result of the paper is a formula for the calculation of the generalized possibilistic risk premium expressed in terms of a utility function and of some possibilistic indicators. Georgescu, I., Kinnunen, J., 2008, Credibility Measures: a Basis for Risk Theory, Turku Centre of Computer Science Technical Report 6/2008. The
paper is an attempt to treat two themes of risk theory based
on the notions of credibility measure and
credibility expected value. The first
theme refers to the definition and analysis of possibilistic portfolios.
With a possibilistic portfolio a probabilistic
portfolio is canonically associated. Risk evaluation in the context of
probabilistic portfolio leads to
an understanding of risk for the
possibilistic portfolio. The second theme deals with risk aversion in an uncertain
situation described by a discrete possibilistic
distribution. Using the credibility expected value we define a
notion of risk premium and we prove a
formula for its evaluation. Heikkilä, M., Kinnunen, J., 2008, Business Risk and Risk Management in Grid Computing. In Carlsson and Störling-Sarkkila, Institute for Advanced Management Systems Research Annual Report 2006, IAMSR Research Report 1/2008. The study discusses the risk
management in Grid computing, where the focus is on the identification of
business risks, attaching them to the relevant context, evaluation of the
possibility of the risks, and acting on them. The paper presents some early
results from the EU-funded research project, AssessGrid. Kinnunen, J., 2004, Price-Earnings Ratio – An Indicator of
Future Earnings or a Red Flag of Overvaluation?, Bank of Finland,
Financial
Markets Department Working Paper 4/2004. The results are discussed in
Bank of Finland Financial Market Report 1/2005, pp. 16-17. The paper discusses the role of Price-Earnings ratio in the market valuation of corporate stocks. The literature of the valuation is extensively reviewed and the historical dynamics and the key driving forces of the P/E-ratio in Finland is under study. The persistence of earnings is rejected and the mean reversion property of the P/E-ratio is observed, which is in line with the international evidence. The paper suggests that the market valuation levels depend on the investors believing that earnings are persistent, i.e., it is believed that the historical earnings are a significant indication of future earnings, which is not supported by observations.
Kinnunen, J., Georgescu, I., 2016, Welfare Effects of Decentralization and Economic Freedoms – a Machine Learning Approach; presentation. Unpublished working paper. Accepted to the 48^{th} Annual Conference of Finnish Political Science Association / XLVIII Politiikan tutkimuksen päivät, March 10-11, Helsinki, Finland. NEW
Decentralization refers to a transfer of authority for public functions from a central government to local governments or to the private sector. Decentralizations often have unintended negative consequences due to various reasons ranging from limited local state capacity/resources to non-local factors. Also, environments can differ greatly and, e.g., copying decentralization arrangements done successfully in a western country to apply them in a developing country may turn out disappointments. We focus on selected consistent political and market decentralizations to get a largest possible comparable coverage of countries. The connectedness of decentralization measures, economic freedom indicators and welfare measures is studied on global level using machine learning and clustering approach. We gather the data from OECD, World Bank and Credit Suisse’s Global Wealth database for about 140 countries. We clean it and drop less meaningful factors to end up with a limited number of clusters. In each cluster indicators go hand-in-hand. Next, we interpret these high-level relationships, firstly, focusing on the rationales found in the literature to explain the successes/failures of decentralizations in different clusters. The results are discussed with respect to endogenous/local and exogenous/non-local factors together with different ideological narratives. Kinnunen, J., Georgescu, I., 2015, Merger and Acquisition Portfolio Selection with Probabilistic and Possibilistic Real Options. Kinnunen, J., 2014, Suomen organisoituva klassinen liberalismi ja poliittinen talous [In Finnish; Finnish Classical Liberalism and Political Economy; presentation]. Poliittisen talouden tutkimuksen seuran Talouden aika -seminaari, May 30, 2014, Turku, Finland. Kinnunen, J., 2014, Economic Knowledge from Misesian Utilitarian, Hayekian Evolutionary and Rothbardian Natural Law Perspectives. Accepted to the 4th International Conference on Power, Culture & Economy 2014, Tampere, Finland, August 25-27. Kinnunen, J., Georgescu, I., 2013, Venture Capitalist Portfolio Selection under Probabilistic and Possibilistic Risk, Accepted to the 20^{th} International Conference on Forecasting Financial Markets, Hannover, Germany, May 29-31, 2013. This
paper views a venture capitalist's (VC's) portfolio selection problem
under two types of uncertainties, i.e., probabilistic risk and
possibilistic risk. A VC faces such a complex situation, when it has
under analysis several potential target companies of which some are
handled in probabilistic terms, while others call for a possibilistic
treatment. A possibilistic treatment is suitable for cases, where
uncertainty is very high. This can be due to lack of available
statistical information, because investment targets are often privately
owned small companies with limited public information, possibly totally
without past sales, without market values, there don't exist comparable
firms to allow comparables-based valuations, and their value largely
depends on intangibles and their strategic future actions. The VC's
portfolio problem corresponds to a situation in which some return rates
on investments are described by random variables, while others by fuzzy
numbers. This paper unifies probabilistic and possibilistic portfolio
selection models resulting in the optimal solution of the mixed
portfolio problem faced by a VC. The paper discusses the added value and
the usefulness of the two-component approach for a VC and presents of a
calculation example with a spreadsheet-tool to demonstrate the
practicality of the approach. Georgescu, I., Kinnunen, J., 2011, Real Option Valuation with Generalized Fuzzy Pay-off Method, Submitted to the 15^{th} Annual International Conference on Real Options, Turku, Finland, June 15-18, 2011. The fuzzy expressions presented in this paper can be used to price real options in a possibilistic setting. The paper extends the fuzzy pay-off method for real option valuation of Collan et al, 2009, by formulating real options models with generalized fuzzy numbers. The possibilistic models become more flexible and can be easily adjusted to obtain the optimal solution. The method provides an intuitive perspective for the uncertainty of the parameters of the models. |