Move your mouse cursor over the table cells below or tap the cell on your touchscreen to see abstracts or other extended information.
You can download the conference program here.
WEDNESDAY - 29. 8. 2018
17:00 – 18:30
Registration, hotel U Kata, Kutná Hora
19:00
OPENING PARTY
THURSDAY - 30. 8. 2018
8:00-9:00
Breakfast
9:10
Conference opening, room A
PLENARY SESSION
ROOM A
INVITED PAPER
Chairman: Joanna Debicka
9:20-10:00
Sixta, J.
The input-output analysis has a long tradition in the former Czechoslovakia starting in the 1960s. These models were originally used for central planning and price re-construction during socialism. Since 1970s, there were three sets of tables, for Czechoslovakia, Czech Republic and Slovak Republic. After 1989, input-output models gained a label of socialist planning instrument and they were mostly abandoned in Eastern countries. On the contrary, in the West sophisticated input-output models became accessible for the most researchers because of the common usage of personal computers. Matrix computations that took months in the past became available for everyone. Alongside with the supply of input-output tables of the EU countries, also researchers in the Central and East Europe became to use these sets of models. In the recent time, input-output models are widely used both on national and international level, ranging from private agencies to the OECD. Nowadays the recovery of Leontief models and theory has gone the Czech Republic, as well. Such models are a standard part of the research for macroeconomic, environmental, regional and similar analysis.
1. SESSION
MACROECONOMIC ISSUES
CHAIRMAN: EMILIA ZIMKOVÁ
ROOM A
10:10-10:30
Cisková, P., Zimka, R.
In the paper, a simple Asada´s macroeconomic model of monetary policy describing the development of the nominal rate of interest and the expected rate of inflation is investigated. The normal equilibrium point of the model is derived and its dynamic stability is studied in conditions of Slovak economy. Questions concerning the existence of business cycles are examined rigorously. The bifurcation equation is found. The formulae for the calculation of its coefficients are gained. Theorem on the existence of limit cycles is stated. A numerical example utilizing the macroeconomic data of Slovak economy from the years 2005-2017 is presented by means of numerical simulations.
10:30-10:50
Kozyra, C.
Transition to democracy and market economy changed the social situation in Central and Eastern European countries after year 1989. After several years economic growth compensated these countries for Initial recession, but increase of inequality was observed, dramatically greater than in the socialistic states. Relationship between economic growth and inequality was analyzed in many studies, but there are different theories, models and conclusions about inequality as the condition and result of economic growth. The most popular concept is Kuznets curve, which assumes that underdeveloped economies experience first growing inequality together with economic growth, but later inequality is decreasing after economy exceeds some level of growth. Recently the most developed Western countries experience inverted Kuznets curve: inequality is increasing again with the economic growth. This analysis is based mainly on time series from OECD databank and from World Bank datasets, because only minority of countries in Central and Eastern Europe belong to the OECD, which delivers the most detailed statistics. Comparison of social welfare level is made with the Sen Index, product of mean income and equality measure (one minus Gini inequality index). The results show that original (not inverted) Kuznets curve seems to be valid for Central and Eastern European countries, although not all countries are on the same stage of development.
10:50-11:10
Kramulová, J., Vincenc, J., Houžvičková, H.,
Margins on buying and selling transactions form one part of this complex system, apart from e.g. transport or trade margins. These financial services in acquiring and disposing of financial assets and liabilities in financial markets constitute an important role in output of financial institutions. Up to now, the Czech Statistical Office recorded only small part of “financial” margins in national accounts, but in the next revision being published in June 2020 the system should include all types.The aim of this paper is to develop missing methodology of capturing margins and demonstrate all difficulties connected with their estimation. Preliminary results are deeply analysed and also included. Our approach is affected by the fact that in the Czech Republic no suitable database with detailed information about transactions is available. That is why simplifications and assumptions needed to be formulated in order to obtain the best possible results. The paper contains time series of cross-border margins, experimental estimates of domestic margins on transactions with securities, shares, investment fund shares and foreign currencies.
REGIONAL ANALYSIS
CHAIRMAN: RICHARD HINDLS
ROOM B
10:10-10.30
Šafr, K., Musil, P., Sixta, J.
The paper is focused on the estimation of symmetric capital matrices by product in the Czech Republic. Symmetrisation is usually related to the intermediate consumption matrix, which is transformed from the dimension product by industry to the product by product matrix that is the starting point for input-output analysis. However, symmetrisation of capital matrix is very rare and has not been done in the Czech Republic. It is a very complex task as it is in fact double symmetrisation as original matrix of stocks of fixed assets is in the dimension type of asset (AN) by industry (NACE). Several data sources were applied the most important are supply and use tables, estimated transformation matrices. The results enable sophisticated input-output analysis focused on the capital and the demand for capital that is caused by economic impulse or shock. Finally, regionalization of the national results is performed by using additional data and RAS procedure. The regional results are compared with other regional economic indicators estimated within the previous research. It extends a list of regional indicators available for analysis and modelling on the regional level.
10:30-10:50
Bartošová, J., Bína, V.
Analysis of incomes of inhabitants is in focus in all developed countries mainly because of the assessment and comparison of lining standards of inhabitants. Knowledge of the income distribution and its comparison using different socio-economic, demographic and spatiotemporal perspectives is a prerequisite for the quantitative evaluation of living standard, level of social welfare and equality in redistribution of the goods created in society. The presented paper focuses on an analysis and comparison of the income distribution shape in the households of all 14 Czech regions. According to the fact that regions of the Czech Republic mutually differ not only in the extent of job opportunities, but also in demographic structure of inhabitants (age, education, etc.), we can expect also differences in the shape of regional income distributions. The aim of the paper is to construct suitable models of income distribution and identify regional differences. For this purpose, both parametric and nonparametric models of frequency distributions will be used. Nonparametric approaches are represented by Gaussian kernel estimates. Among parametric methods we can employ some of simple probability distribution models – most frequently two- and three-parametric lognormal models – or chose more demanding but also more accurate methods, i.e. estimate a model of income distribution based on finite mixtures of densities. This method is usually used for the modeling of distributions of random variables in heterogeneous populations and therefore in our case it comprises a better alternative.
10:50-11:10
Šafr, K.
Since 2008, when the Czech Statistical Office stopped publishing the territorial structure of Czech regional export, there has not been any other source of these data. This paper focuses on methodology of estimating Czech territorial export at the regional level through a combination of two other data sources: Czech regional input-output tables and world input-output tables for the year 2013 in current prices (in millions of dollars). These two data sources are combined by an adjusted cross-entropy method and other information provided by official statistics. The final figures showed that the overall structure is similar at the regional level. However, the regional detail differs between regions and there are some special examples of connections, which are not as strong in other regions. Moreover, the presented paper is strongly connected with the methodology supporting the data sources. The presented methodology can be used for estimating model application data, e.g. computable general equilibrium and others.
11:10-11:40
Coffee Break (ROOM A)
2. SESSION
SOCIAL ECONOMICS ISSUES
CHAIRMAN: JAKUB FISCHER
ROOM A
11:40 – 12:00
Biernacki, M.
The quality of national healthcare systems in Visegrad Group countries (G4), was analyzed based on data obtained from Eurostat, WHO, and Health Consumer Powerhouse for 2004 and 2012 and compared to selected EU countries. The non-parametric DEA method was applied to analyze efficiency and the standardized result of the Euro Health Consumer Index was employed to analyze effectiveness. Relying on these two measures, the performance of national healthcare systems was assessed by means of the “modified” Pareto ordering. An index which takes into account 4 dimensions: efficiency, effectiveness, patients‘ satisfaction and social welfare growth was proposed for system quality assessment. In the analyzed period, the performance of social healthcare systems in G4 countries increased the most in Slovakia (by 23 percentage points), Hungary (21 points) and the Czech Republic (8 points). In Poland the system performance decreased by 1 percentage point. The quality of healthcare systems in G4 countries is higher than in Portugal and Spain and comparable to the system quality in the United Kingdom and Germany when we treat each dimension equally. When the quality of the system is measured from patients’ point of view (performance and patients’ satisfaction weigh more), G4 countries offer a significantly lower quality of healthcare systems.
12:00-12:20
Hužvár, M., Kaščáková, A.
Measuring the well-being of people and the progress of societies should be a key priority for development and policy making. Better Life Index is a newly designed tool which provides a comprehensive picture of well-being in OECD countries and other major economies, by looking at people’s material conditions and quality of life across the population. The index is based on twenty-four indicators grouped in eleven topics concerning housing, income, jobs, community, education, environment, civic engagement, health, life satisfaction, safety and work-life balance. However, the OECD does not rank countries by the Better Life Index. Instead, users are encouraged to assign and share their own weights for the topics to design a composite indicator that can express their preferences on the contribution of each topic to well-being. The aim of this paper is to present and discuss the “benefit of doubt” approach to the construction of a composite index of well-being based on the eleven topics. By this approach, we assign to each country individual weights for the topics that can emphasize the strengths of the country in people’s material conditions and quality of life. We show that such composite index is capable to provide policy makers with additional valuable information on untapped potential of individual countries in well-being.
12:20 – 12:40
Girul, A., Mazurek, E.
Increased numbers of disabled persons both in Poland, as well as in other European countries, have been triggering many political debates. Guaranteeing dignity and full inclusion of persons with disabilities in society is one of the main objectives embraced by the new disability action plan 2016–2020 of the Council of Europe as regards the promotion of full social participation of persons with disabilities. The accomplishment of the targeted objective depends on an insightful monitoring of the situation of persons with disabilities. Factors that differentiate persons with disabilities with respect to their economic activity have been identified focusing on the problem of employment of disabled persons. For this purpose a multivariate correspondence analysis was applied with data provided by the National Census of Population and Housing 2011. In Poland, women aged 65+, with secondary education or lower, predominate the group of persons with disabilities. Most of disabled persons are related by marriage contract. Statistical analysis did not indicate any significant correlation between the categories of economic activity, and marital status and the degree of disability, whereas significant correlations have been observed between the categories of economic activity and such characteristics as: age, sex, education, and pursued profession.
12:40-13:00
Marciniuk, A., Zimková, E.
Demographic trends in Europe reveal that the pension funding gap will become one of the key social issues in coming years. On the other hand, many of these people hold a large amount of wealth in their property which, being reasonably utilized by equity release products, could help them cover their needs. People can surrender their real estate to a company interested in the acquisition of their property in exchange for the whole life monthly benefits. The aim of the contribution is to analyze the potential benefits of marriage reverse annuity contracts in the Slovak Republic by the use of the Svensson model function, considering the reversionary annuity and the real value of estate in different cities. The real value of the properties is determined by the place where spouses live and has significant influence on the amount of the benefit. While in many countries the equity release products have been offered to clients for dozens years, a product of this nature has not been established so far in the Slovak Republic. Hopefully, this contribution might initiate discussions on the introduction of similar equity release products market in the Slovak Republic as well.
MULTIVARIATE STATISTICAL METHODS
CHAIRMAN: PAVOL KRÁL
ROOM B
11:40-12:00
Kráľ, P., Lesáková, L., Gundová, P.
When studying key factors and barriers determining innovation activities of enterprises, we need to identify unique categories of enterprises which provides us with a good representation of innovation focused behavior of enterprises. Based on their innovation activities, enterprises are usually classified into two basic categories: innovative enterprises and non-innovative enterprises. Innovation activities of an enterprise could result in various types of innovations – product innovations, process innovations, organizational innovations and marketing innovations. These types of innovations could be combined for each enterprise in very different and unique way, which results in high variability of differences among enterprises. Consequently, standard classification into two groups seems to be unnecessary rough one. In our contribution, we aim to identify a more suitable set of innovation categories by applying various clustering methods to data of Slovak medium and small enterprises collected in the period November 2015–January 2016. Identified innovation categories will be examined more closely via descriptive statistics and exploratory graphical techniques.
12:00-12:20
Řezanková, H.
Cluster analysis is a useful statistical tool for data exploration. It can help to identify groups of similar objects (e.g. countries) according to selected variables (e.g. economic indicators). The created groups (clusters) can be characterized based either on the variables used in clustering or on some other variables. The problem with using the methods of cluster analysis is that the analyst can obtain different results (assignments of objects into clusters) by different methods, and moreover, he needs to determine the number of clusters. Many coefficients for solving this problem have been proposed until now. However, the suggestion obtained by a certain coefficient can differ from another suggestion obtained by another coefficient. In addition, values of a certain coefficient can differ depending on the implementation in the software product. In this contribution, different approaches to the silhouette coefficient calculation are discussed. They concern the implementations of calculations in the R language and IBM SPSS Statistics system. It is analyzed how the different results influence decisions of an analyst in terms of both the choice of the suitable assignment of objects into clusters obtained by different algorithms, and the determination of the suitable number of clusters. The studied problems are illustrated using selected methods of cluster analysis applied to the EU countries characterized by the gender indicators.
12:20-12:40
Cibulková, J., Šulc, Z.
Cluster analysis is a multivariate statistical method with a wide range of possible applications. It is especially useful for market segmentation, in which objects are divided into homogenous segments (clusters) which are further analyzed to obtain segment-specific insights. This contribution presents an application of cluster analysis on multivariate data provided by a company from a field of tourism. The dataset contains information about 5,755 travels of its customers, such as the number of passengers traveling together, age, nationality, route details, price, number of destinations booked, etc. The goal of the analysis is to divide the customers into several distinct segments according to their profiles, while demonstrating the importance of distance measure selection and linkage method selection. The results of the analysis will help to develop a targeted marketing program for the company. Since the dataset contains categorical or categorized variables, hierarchical cluster analysis for categorical data is applied to perform the market segmentation. Due to the fact, that clustering process is always strongly dependent on a similarity measure used and also on a linkage method, the optimal cluster assignment is being chosen among five similarity measures for categorical data and three linkage methods. Clustering solutions corresponding to a specific similarity measure and a specific linkage method are compared and evaluated by internal evaluation indices, which allow finding the optimal number of segments, evaluate their internal consistency and determine the best clustering solution possible.
12:40-13:00
Stachová, M., Sobíšek, L.
One of the main task in each analysis of companies’ financial status is to correctly define the criteria that can describe the financial health or financial distress of these enterprises. In general, the financial distress is a situation in which company cannot pay or has difficulty to reach its financial obligations. Our data set consists of three financial indicators of Czech enterprises. These longitudinal data are collected over a few consecutive years. We applied the model-based partitioning and the K-means partitioning to these longitudinal data to cluster the time trajectories of these criteria and subsequently we compare the accuracy of these algorithms. We use packages “mixAK” and “kml” of the statistical system R in our analysis.
SESSION 3
INSURANCE AND DEMOGRAPHY
CHAIRMAN: CYPRIAN KOZYRA
ROOM A
14:40-15:00
Špirková, J., Kollár, I., Szűcs, G.
The Act 43/2004 Coll. on the Old-Age Pension Saving Scheme and on amendments and supplements to certain laws offers, among others, an old-age pension product with increasing annuity payments. In our model we consider a monthly paid annuity increasing yearly in geometric progression. Just the determination of the geometric progression quotient is a crucial task, especially when we consider valuation of the future cash-flows based on risk-free bond yield curves. On the basis of actuarial formulas we point out on a mutual relationship between the risk free bond yields and the yearly increase rate of the pension payments. In our contribution we also examine the impact of the increase rate on the overall benefit of the pensioner.
15:00-15:20
Zmyslona, B., Marciniuk,A.
European societies are affected by the aging of society. Pensioners in post-communist countries constitute a significant percentage of people at risk of poverty. In addition, such people more often than others suffer from chronic diseases. These phenomena cause a significant decrease in the quality of life of these people. The article considers a proposal for financial and insurance products that can increase the financial security of pensioners. These products can provide an additional source of pensioners‘ income. In addition, funds obtained through the new products can be used to subsidize the costs of treatment. Co-payment in case of chronic diseases can significantly improve the quality and effectiveness of treatment.
In our paper, we propose a new combined financial and insurance product which consists of marriage reverse annuity contract and dread disease insurance for a husband and for a wife. Marriage reverse annuity contract is an equity release products offered to elderly spouses. Owners receive annuity-due benefits in return for the transfer of the ownership onto the company (mortgage fund) and they have an ensured right to live in property until their death. The analyzes connected with dread disease insurance are carried out on the basis of data from Lower Silesia regarding lung cancer, in particular morbidity, course of disease and mortality.
15:20-15:40
Fiala, T., Langhamrová, J.
To ensure appropriate pension for seniors is a very important task for social policy. A very frequent discussed consequence of the population aging is the financial sustainability of the pension system, especially, if it is based on PAYG principle. One of the possible measure supposed is logically increase of the retirement age. There naturally arises an important question of the appropriate linkage of the retirement age with the level of mortality. Alternative threshold can be based on the concept of so-called prospective age (the age in which the remaining life expectancy is the same as the remaining life expectancy of a person in given age and standard mortality pattern). However this threshold does not reflect well the actual age of the person and it would cause that the period of expected economic activity would rise while the period of pension receipt would remain constant. A possible compromise of these both definitions may be the concept of so-called “relative” prospective pension age based on the relative remaining life expectancy (the ratio of remaining life expectancy and the total expected life span – the sum of age reached and remaining life expectancy at this age). The pension age can be thereafter defined e.g. as the relative prospective age corresponding to the age of 65 years and mortality pattern of European Union in 2016. The paper presents the computation of proposed pension age in the Czech Republic, Slovakia, Poland and Spain until 2080 based on the relative prospective age concept and corresponding old age dependency ratios. The baseline scenario of the population projection of Eurostat 2015 has been used for mortality projection.
15:40-16:00
Sotona, P.
The article focuses on the mortality risk assessment in the insurance industry for the Czech Republic. New accounting standard IFRS 17 requires to disclose confidence level at which the insurance company assessed insurance risks inherent in issued insurance contracts. This article analyzes mortality risk which can be further split into four subrisks: volatility, catastrophic risk, level uncertainty and trend uncertainty. On the practical example of insurance portfolio with term insurance product I present the application of various statistical methods to assess mortality risk and to estimate total risk adjustment (under IFRS 17) for mortality risk on 90% confidence level. Final results are analyzed and commented with conclusions.
16:00-16:20
Mazouch, P.
Young adults‘ excess mortality – between age of adolescence and young adult – is well known fact. In some literature this excess is called „mortality bump“ and reason why we can find it in the development of mortality is usually justified just theoretically and measured for males only. This article describes development of the „bump“ in cohort perspective of males in the Czech Republic. In time we can observe changes and shift of the excess to the lower level for younger generations. It seems that younger cohorts have lower mortality risk but the question is if the increase in the period of 18-21 years is decreasing also or not.
TIME SERIES ANALYSIS METHODS
CHAIRMAN: LUBOŠ MAREK
ROOM B
14:40-15:00
Flimmel, S., Fojtík, J., Malá, I., Procházka, J.
Autoregressive process AR(p) is very popular and frequently used when working with time series, especially in financial mathematics. One of the requirements for working with AR(p) is the ability to estimate parameters of the model correctly. However, we currently often deal with big data, which can lead, among others, to a higher probability of outlier presence. As it is known, standard methods for parameter estimation are often not able to work correctly with outliers, and, consequently, standard estimates are usually biased. Therefore, working with sufficiently robust methods has increased in importance. In this paper, we present several robust moment methods for parameter estimation in AR(p) and we compare them using a simulation study. Outliers in the simulations are modelled using two most frequently used outlier models: additive outlier (AO) and innovative outlier (IO). For the simulation study, we use the R statistical software.
15:00-15:20
Blatná, D.
When analysing labour market trends, the employment rate serves as a key social indicator, defined as a ratio measuring the proportion of the employed working-age population. The rate of employment is one of the headline indicators in the Europe 2020 strategy for smart, sustainable and inclusive growth. The development of the employment rate both in the EU and the Czech Republic in 2005–2016 is examined with reference to achieving the objectives set in the Europe 2020 strategy, the Czech Republic having already exceeded the 2020 national target. The aim of the paper is to analyse the relationship between the employment rate and selected socio-economic indicators such as real per capita GDP growth rate, inflation rate, educational attainment, level of lifelong learning, social benefits, etc., in the Czech Republic. The regression analysis was used as the main tool of analysis. The employment rate regression analysis was carried out applying multivariate non-stationary time series methodology. The analysis covers the period between 2005 and 2016.
15:20-15:40
Witzany, J., Fičura, M.
The aim of this paper is to propose and test a novel PF method called Sequential Gibbs Particle Filter allowing to estimate complex latent state variable models with unknown parameters. The framework is applied to a stochastic volatility model with independent jumps in returns and volatility. The implementation is based on a novel design of adapted proposal densities making convergence of the model relatively efficient as verified on a testing dataset. The empirical study applies the algorithm to estimate stochastic volatility with jumps in returns and volatility model based on the Prague stock exchange returns. The results indicate surprisingly weak jump in returns components and a relatively strong jump in volatility components with jumps in volatility appearing at the beginning of crisis periods.
15:40-16:00
Piotrowska, M., Kośny, M.
The paper is aimed at explaining differentiation in economic security of households through cohort differences. Identifying cohorts is based on common educational and professional experiences. The research introduces the concept of economic resourcefulness defined as the capability to make economic decisions that contribute to economic security of a household. Economic security of a household is defined as the ability to achieve income necessary for covering household needs at its suitable level and to create financial reserves to be at disposal in case of unfavorable accidence. The research uses an exploratory analysis based on structural equation modeling. The questionnaire survey carried out in Poland in 2013 is a source of data for observed variables in an empirical model of economic security. The findings confirm that common educational and professional experiences, controlling for economy conditions, influence economic resourcefulness of household’s members and, as a consequence, their economic security. The conclusion, however, refers to the Polish society and socio-economic development experienced by cohorts of Poles since 1990.
16:20-16:50
Coffee Break (ROOM A)
4. SESSION
INSURANCE MARKET
CHAIRMAN: BEATA ZMYSLONA
ROOM A
16:50-17:10
Gubalova, J., Medveďová, P., Špirková, J.
About fifteen to twenty years ago, funds invested in a life insurance in Slovakia were being valued by a guaranteed so-called technical interest rate of (5-6)% p.a. Gradually, this interest rate was declined to 0.7 % p.a. in 2016. Consequently, with the entry into force of the Directive2009/138/EC of the European Parliament and of the Council of 25 November 2009, known as Solvency II, and the Directive 2014/51/EU of the European Parliament and of the Council of 16 April 2014, known as Omnibus II Directive, insurance companies should guarantee maximum returns on finance on the level of the yields of the risk-free bonds, traded in the eurozone. This paper offers a case study related to the amount of premium, the advantages or disadvantages of an insurance policy and of the selected insurance contracts. Based on our analysis, we recommend insurance as an actual risk coverage and not as an investment.
17:10-17:30
Fojtík, J., Procházka, J., Zimmermann, P., Švehláková, M.,Macková,S.
Accurate modelling of insurance liabilities which is able to reflect the time value of financial options and guarantees (FOGs) is one of the essential actuarial tasks required by Solvency II and IFRS. Liability modelling of large insurance portfolio that requires running thousands of economic scenarios is very demanding on computational time. In our previous research, we proved that cluster analysis, which limits the number of the modelpoints to be modelled, is a good approximative method which can be used to speed up the liability estimation while preserving the accuracy. Using a faster method for the estimation of the liability value allows the actuaries to study more investment strategies and provide a more complex analysis or sensitivities to understand market risk or manage proper investment strategies by asset liability management. In this paper, we focus on the application of the dynamic asset liability management in life insurance business and the faster liability model based on cluster analysis. In the results, we present the answer to whether the faster liability model is suitable for two basic ALM methods – cash flow matching and duration.
17:30-17:50
Debicka, J.
The insurance premium can be decomposed into the savings premium and the risk premium. The savings premium represents the increment in the reserve, that is a fund covering the increased risk of the insured event in the future. Whereas, the risk premium represents the expected cost which is related to the actual chance of an insurance event occurring in the future. A critical illness insurance allowing period premiums paid by both healthy and ill person (e.g. not terminally yet) is considered. The aim of the paper is to analyze the influence of different kind of premiums on the value of net premium components for each year of the insurance period.
17:50-18:10
Sládek, V., Malá, I., Čabla, A.
Insurance company has to estimate its liability (present or future) from its business. Estimation could be based on parametric or non-parametric approaches. Correct identification of probability distributions in parametric approach could lead to accurate inferences about liability. It could lead to greater reliability in estimation and in the best case, to reduce the costs of capital. Skewness-kurtosis graph is a very useful tool for an identification of probability distribution. It is based on intra-relations between moments of the probability distribution. Probability distributions typical for insurance are lognormal, Pareto, gamma distribution and many others. A lot of them could be classified as heavy-tail or long-tail distribution, i.e. extreme values have high probability of being selected in a random sample. Estimates of product moments of skewness and kurtosis are sensitive to the extreme values presence and are limited by a sample size. L-moments are a robust alternative to product moments. This contribution is focused on skewness-kurtosis graph based on L-moments, so-called L-skewness, and L-kurtosis. Its application is shown on Monte Carlo simulation and also an empirical study is presented. A distribution of the amount of Motor third-party liability claims is identified.
APPLICATIONS OF STATISTICAL METHODS IN ECONOMY
CHAIRMAN: MARTIN BOĎA
Room B
16:50-17:10
Marek, L., Vrabec, M., Berka, P.
Our paper deals with the ex-post verification of models of wage distributions designed to predict wage distributions in the last three years. We will use the prediction results of Lognormal, Lognormal (3p), Johnson SB, Log-Logistic, Log-Logistic (3p) and Normal Mixture distributions and compare them with the empirical distribution from the period 2015-2017. The selection of the used distributions is based on the wage distribution models for the years 2000-2014. Our results show, that the best (and comparable) results can be obtained using three-parameter Log-logistic distribution and Normal Mixture distribution with two components. These results confirm our expectation that due to the fact, that empirical wage distribution becomes less smooth over time, a mixture model should be preferred for the future.
17:10-17:30
Král´, P., Laco, P.
World Wide Web, especially corporate websites are used by businesses to publish contacts, billing and service information, and product offerings. Businesses can also engage in interactive communication through questionnaires, discussion forums, and on-line news to help them react very flexibly and provide feedback to their customers. In practice, however, not all businesses use these options effectively and the level of quality on business websites varies wildly. Consequently, a flexible and easy applicable methodology providing companies with a possibility to evaluate their own presence would be beneficial. There are many mathematical, statistical and decision-making methods which could form its core. In our contribution, we create such a methodology for scoring of corporate websites based on partial least squares path modeling. The proposed methodology is intended for application in the conditions of Slovak SMEs.
17:30-17:50
Grausová, M., Hužvár, M., Rigová, Z.
The implementation of the Higher Education Act of 2002 in Slovakia has significantly changed the economic conditions for universities that were transformed to public higher education institutions. Subsidies from the state budget are allocated to individual universities based on their share on the total performance of all public higher education institutions in education and research. The competition for financial resources has greatly influenced the behaviour of universities. Although some effects of this system of financing are considered questionable, it is naturally expected that the increasing competition may have a positive impact on the productivity of universities in the transformation of input resources into desirable outputs. We study the productivity change of Slovakian universities over the period when the new legislation is applied. Malmquist index is used for the assessment of total productivity change, and data envelopment analysis for calculating efficiency scores to construct the index.
17:50-18:10
Gardon, A.
19:00
Conference dinner
FRIDAY – 31. 8. 2018
8:30-9:30
Breakfast
5. SESSION
FINANCIAL MARKET
CHAIRMAN: EDYTA MAZUREK
Room A
09:30-09:50
Boďa, M., Kanderová,M.
The paper investigates usefulness of a rebalancing strategy that conjoins polar investment styles in both asset pre-selection and portfolio selection. It is customary in investing to adhere to a particular investment style defined on the basis of a measurable quantifiable characteristic (using screening) and to invest into assets that have this characteristic high or low, respectively. The strategy assumes that the portfolio is created by investing into both such polar classes of assets in certain proportions and these proportions would be maintained by periodic and threshold rebalancing over the investment horizon. The usefulness of the rebalancing strategy is evaluated by a case study oriented on the US stock market.
09:50-10:10
Virdzek, T., Kubaška, P., Cisková, P.
Modern portfolio theory represents the approach to constructing an optimal portfolio based on the weighting of individual assets in the portfolio. We can call this approach as an active, because the investor assigns higher weights to the assets that are able to achieve higher performance on the basis of past performance. In addition to the active investment approach, we also know the passive investment approach, through which the investor seeks to evenly distribute capital among other assets (naive diversification). The aim of this paper is to evaluate the performance of the active approach to weighting in the portfolio versus naive diversification. For the purpose of this work, we divide the data sample into two parts. The first part will serve to determine weights in the portfolio (training set) and in the second part we will test the performance of this portfolio versus the approach of naive diversification (testing set).
10:10-10:30
Ptáčková, V., Štěpánek, L., Hanzal, V.
Economists, politicians or managers always want to know what mood the economy is in. Economic surveys, Short-term statistics and Business tendency survey, can give the answer to this question. However, the short-term business statistics describe real data about companies; the results are published with a two months delay. By way of contrast, the Business tendency survey indicates predictions of this economic development. Since the Business tendency survey is based on opinions of selected companies, a quality and confidence of predictions in the previously mentioned survey is opened to discussion. The article focuses on evaluation predictions and the real development of industrial companies in the Czech Republic. Authors are comparing results from Short-term statistics in the industry and the Business tendency survey. The Czech Statistical Office collects data from the both surveys. Using mathematical and statistical methods, authors want to scale and categorize real data-based economic development, recorded by short-term business statistics, into categories corresponding within predictions, offered by Business tendency survey, such that the appropriate categories from the both sources correlate themselves enough. The categorization attempt is performed for real data of two economic indicators – employment and sales. In case the categorization of real data is successful, Business tendency surveys published without any significant delay can provide sufficient and early idea about near future economic development. Furthermore, the results of the analysis can help to understand respondents´answer in the Business tendency survey and possibly update methodology, will be used to assemble indicators of the trust.
10:30-11:00
Coffee break (room A)
6. SESSION
STATISTICAL METHODS IN ECONOMY; HISTORY OF STATISTICS
CHAIRMAN: PETER LACO
Room A
11:00-11:20
Vozár, O.
Randomised response (RR) techniques are used to get reliable estimates on variables of sensitive nature. Unlike of estimation of a population proportion there are just few RR techniques to estimate a population total of quantitative. Antoch et al. (2002) proposed a technique which provides high degree of protection of sensitive respondent characterics. Instead of revealing true value of a sensitive variable surveyed, the respondent answers whether it is larger (smaller) than a number which he selects at random. The number is not known by the interviewer. This technique can be applied in the most common sampling plans to construct unbiased Horvitz-Thompson estimator of a finite population total. We present two modifications of the technique by Antoch et. al. (2002) First we study case that interviewer knows the randomly selected number. The second case deals with situation, that it can be sensitive for respondent to report both small and large values of a sensitive variable (e.g. income). Depending on the value of randomly selected number, respondent answer different questions to be embarrassed as few as possible. Results of simulation study illustrating the performance of proposed estimators is included.
11:20-11:40
Heilpern, S.
The contribution is devoted to the risk process in which the claims amount are uncertain. The uncertainty is modeled using the randomness and fuzzines simultaneously and the claim amount are treated as the fuzzy random variable. The definition of the fuzzy random variables proposed by Kwakernaak (1978, 1979) is used in this contribution. The problem connected with the ruin of such process is investigated and some numerical examples are presented. The situation when the initial capital is equal zero is studied and the fuzzy ruin and the mean value of it are computed for the fuzzy exponential random variables. The spread of fuzzy ruin, the measure of uncertainty, is investigated.
11:40-12:00
Závodský, P., Šimpach. O.
Land Statistical Office (LSO) in Prague started its activity 120 years ago. In year 1898 and year 1919 it was overtaken by State Statistical Office of new Czechoslovakian state. During the first two decades the number of workers of statistical chair of LSO increased almost four times and its scientific and publication activities significantly broaden. Land Statistical Office annually published (in Czech and German) several volumes of Zprávy / Mittheilungen (Reports) that contained mainly the results of analysis of meteorological observations and data of agricultural statistics and also the results of land self–government statistics, statistics of inhabitants, education and several branches of economy statistics (all for Bohemia). Methods of agricultural statistics and statistical graphs used by LSO, evoked recognition at the Conference for land statistics in the Austrian part of the monarchy and at meetings of the International Statistical Institute (ISI). Future university professors D. Krejčí, K. Engliš, V. Mildschuh, F. Weyr and W. Winkler underwent statistical practice in LSO similarly as the leaders of Czechoslovakian state statistics D. Krejčí, F. Weyr, J. Auerhan and J. Mráz.
12:00-12:20
Conference closing
12:30-13:30
Lunch
14:00
Guided tour to the historical centre of Kutná Hora
20:00
Dinner
SATURDAY – 1. 9. 2018
8:30-9:30
Breakfast
10:00
The full-day walking trip will be situated to the southern surroundings of Kutna Hora. During the 18km long trip we will see Vrchlice valley (with mills, nice views on Kutna Hora panorama and waterfalls), „Friendly Brewery“ in Malesov founded in 2017, Roztez Chateau well-known from some movies with the majestic English park, and finally the ruins of Hussite Sion Castle. Return by the local train. There are some possibilities to shorten the trip.
19:00
Dinner
SUNDAY – 2. 9. 2018
8:30-9:30
Breakfast
10:30
Check out and departure