### EUROPEAN COUNTRIES ANALYSIS

USING ROBUST REGRESSION METHOD

##### Dagmar Blatná

Prague University of Economics

Czech Republic

European countries can be characterised by indicators of general economic background, employment, inovation and research, science and technology. Values of these indicators are varying among European countries. The most used statistical tool for analyzing dependences is regression analysis. The classical statistical approach – the least squares method (LS) may be highly unsatisfactory in the presence of outliers which can be supposed in analysis of European countries data. In such case robust regression is acceptable and useful tool.

### THE CAPITAL ASSET PRICING MODEL OVERVIEWED

##### Martin Boďa, Mária Kanderová

Matej Bel University in Banska Bystrica

Slovak Republic

One of the dogmas of contemporary financial risk management comprises the assumption that the returns of an asset can be modelled as linearly dependent on risk-free rate and the resulting Capital Asset Pricing Model (CAPM) is received great popularity. The contribution focuses upon the CAPM and discusses its appropriateness, with special attention being given to stability of its parameters over time.

### ON TWO VERSIONS OF THE HARTWICK´S RULE IN AN EXHAUSTIBLE RESOURCE MODEL

##### Anton Dekrét

Matej Bel University in Banska Bystrica

Slovak Republic

In this contribution two versions of the Hartwick´s rule, the one with-the other without auxiliary (evaluation) variables, are treated and compared in an exhaustible resource model. First the geometrical backround of auxiliary variables is described and then these versions are compared both in the control and in the optimal control model with the conditions of competivity and equity.

### INCREASING LEVEL OF EDUCATION IN SELECTED CENTRAL EUROPEAN COUNTRIES: A CERTAIN ADVANTAGE?

##### Savina Finardi, Jakub Fischer, Petr Mazouch

Prague University of Economics

Czech Republic

Educational structure of the population has been indisputable changed in many post-communistic central European countries in the last ten years. This fact is influenced by the effort of the governments to the convergence of the ratio of tertiary-educated people to the ratios in the old European countries and OECD countries.

It is a question, if this process pointed to the increasing number of graduates and the increasing ratio of tertiary-educated people is certainly positive. At first sight it is clear. Highly educated people should work in branches with higher value added, should have higher labor productivity and higher wages and so on. But, from other point of view, it is necessary to compare this process with the demographic situation in these countries. It is obvious that the birth-rate has decreased in recent years and the youngest part of population is smaller. With relation to the growing number of students in tertiary education system and the number of graduates, the ratio of graduated people rises too rapidly. It could cause some troubles based on over-qualification.

The aim of the paper is to compare the trends in tertiary education systems in the Czech Republic, Slovakia and Poland in the context of the general demographic trends in these countries. Relation to the main economic indicators is also considered.

### ERROR CORRECTION MODELS AND REAL CONVERGENCE – CASE OF SLOVAKIA

##### Rudolf Gavliak, Vladimír Úradníček, Emília Zimková

Matej Bel University in Banska Bystrica

Slovak Republic

On January 1, 2009 the Slovak Republic enters the Economic and Monetary Union. This step is is an aproval that the country has fulfilled nominal criteria for monetary integration. The main aim of the monetary integration is sustainable economic growth and convergence of the real economy to most advanced EU countries. Long lasting mutual trend of analysed indicators is searched by cointegration. In a short term there might be some deviations which are studied by error correction models (ECM). The output of the contribution is twofold. Firstly we will analyze the long-lasting equilibrium of the real economy and secondly we will estimate the duration whithin which the searched variables are returning to the long-lasting equilibrium in the case of the short-term deviations.

### HOW TO TRANSFER BETWEEN DIFFERENT QUALITY MEASURES OF CLASSIFICATION MODELS

##### Lubomir Hanusek

Prague University of Economics

Czech Republic

There are many ways how to evaluate a performance of classification models. The most known statistics are Gini Coefficient, Lift, Kolmogorov – Smirnov Statistic, Sensitivity, Specificity or Percentage of Correctly Classified Cases. You may have a favorite measure that you understand perfectly. But it can happen that you read a scientific article, documentation of a model or a statistical textbook where an author uses a different measure that you can hardly imagine because you are not familiar with it. It would be nice to have a tool able to make a transformation to your favorite measure. Are we able to make such transformation, under what conditions? How difficult is it to calculate?

### PROCESSES WITH DEPENDENT RISKS

##### Stanisław Heilpern

Wroclaw University of Economics

Poland

The paper is devoted to the risk processes when the assumption of the independence is omitted. The some models are presented. First, the case when the frequencies are dependent in the multiclass models are studied, and second, the processes with dependent claims are investigated. The two class process with dependent pairs of risks is also studied. The probability of ruin and the influence of the degree of the dependence on such probability are investigated. The copulas are used to this end. The application to the modeling portfolio credit risk is mentioned.

### HEALTH-ORIENTED LIFESTYLE RESEARCH

##### Cyprian Kozyra

Wroclaw University of Economics

Poland

The main aim of this paper is presenting results of survey research of such issues:

1) awareness of influence of lifestyle on health,

2) assessment of practiced lifestyle.

Population of concern were students of Wroclaw University of Economics and sample was selected by means of random stratified cluster sampling.

Lifestyle is defined as the all behaviours chosen in aim to mould own life. These detailed questions were investigated in research: What do respondents mean about healthy lifestyle? Are they aware of influence of lifestyle on health? How do they assess their lifestyle and why? Do they feel, that their lifestyle should change and how to do it? What is the main respondents’ goal of practicing healthy lifestyle?

Results of research give positive image, but rather nobody knows whether it is the sign of reality or the sign of respondents’ expectations of their life.

### ASPECTS OF POLISH ENTREPRENEURSHIP

##### Marek Kulesza

Jan Dlugosz University in Czestochowa

Poland

Entrepreneurship is considered as the main condition of social-economic development and as key factor to achieve competitiveness in scale of entire economy. This paper presents characteristics of the Polish enterprises, their innovativeness and their institutional environment. Position of Poland in cross-national assessment of entrepreneurial activity due to Global Entrepreneurship Monitor (GEM) and other entrepreneurship surveys and place of Poland in international rankings of competitiveness are also analyzed.

### INFLUENCE OF SEASONALITY ON ACCOMMODATION FACILITIES’ CAPACITY PLANNING

##### Peter Laco

Matej Bel University in Banska Bystrica

Slovak Republic

This paper is concerning seasonality in tourism, especially in accommodation facilities. The possibilities of various methods application in analysis of seasonality in Slovak accommodation facilities are discussed. Concrete application of certain mathematic and statistic methods in hotel management is also attached. Application of time series analyse on Slovak accommodation facilities’ data shows us character of seasonality in various kinds hotels.

We focus on accommodation facilities’ capacity planning, especially in Banska Bystrica region. We present results of survey that show us an alternative view to data collected by Statistical office of Slovak republic. We also present measures to reduce negative effects of seasonality on accommodation facilities’ capacity usage.

### FLUCTUATIONS IN A TWO-REGIONAL MODEL

WITH FIXED EXCHANGE RATES

##### Peter Maličký, Rudolf Zimka

Matej Bel University in Banska Bystrica

Slovak Republic

T. Asada introduced in Discussion Paper Series No. 44, Chuo University, Tokyo, Japan, 2003 a two-regional nonlinear macrodynamic model which describes the development of income, capital stock and money stock in two regions connected through inter-regional trade and inter-regional capital movement with fixed exchange rates. The model can be aplicable to the analysis of the dynamic interaction of two countries in the period of their fixed exchange rates. The model can be also used to the analysis of dynamic interaction of two countries belonging to Eurozone, as the currency integration is the extreme form of the system of fixed exchange rates. The Asada´s model is a five-dimensional system of nonlinear differential equations. Asada analysed in his paper the question of stability of the equilibrium of the model. In the presented contribution the question of the existence of business cycles and tori in the model is solved. Numerical examples illustrating gained results are given.

### MEASURING INEQUITY

IN POLISH INCOME TAX SYSTEM

##### Edyta Mazurek, Marek Kośny

Wroclaw University of Economics

Poland

In this paper equity in income taxation is defined by means of three axioms, introduced by Kakwani and Lambert (1998). Tax system is equitable if all axioms are satisfied. Violation of them – by an income tax system – produces negative influence on the redistributive effect of the tax. This negative influence provides the means to characterize the type of inequity present in an income tax system. We analyze these effects, using decomposition of the redistributive effect.

Actual redistribution depends both on construction of tax schedule and unintended effects, such as re-ranking of incomes, caused by taxation.

To separate both components, several decompositions of redistribution index have been described. The first one, proposed by Kakwani (1984), was further developed by Aronson, Johnson, Lambert (1994) and van de Ven, Creedy, Lambert (2001).

They assumed division of the whole population into groups with exactly equal incomes. Attempts of application (see Urban, Lambert, 2008; Vernizzi, Pellegrino, 2008) indicated necessity of generalization – models failed to capture re-ranking effect. This failure forced abandonment of the idea of exact equal incomes. And caused the problem of choosing the optimal bandwidth – what incomes could be classified as “close enough” to be put into the same subgroup: obtained results (see Vernizzi, Pellegrino, 2008) showed strong dependency of income interval on decomposition results.

Therefore, we present a measurement system and an analysis of redistribution capability of the tax system. All analyses are made basing on data from the tax offices.

### TOWARDS INFORMATION-BASED WELFARE SOCIETY

##### Walenty Ostasiewicz

Wroclaw University of Economics

Poland

After collapsing the nazi war-state during successive 30 years there were formed in Europe welfare states. They were formally based on the social contract between factories and trade unions .

Globalization, which emerged in 1980s, destroyed this social order. A new form of social arrangement are discussed by various international organizations and institutions. The mort important of them are highlighted in this paper.

### ASSESSMENT OF POPULATION DIVERSITY AND HETEROGENEITY

##### Stanisława Ostasiewicz, Walenty Ostasiewicz

Wroclaw University of Economics

Poland

All Men are created equal, that they are endowed by their Creator with certain unalienable Rights, that among these are Life, Liberty and Pursuit of Happiness. So is declared in Declaration of Independence as for ago as 1776.

Nowdays we know that among equal there are more equal and less equal. From HDR2005 we learn, for example, that the world’s richest 500 individuals hare combined income greater than that of the poorest 416 000 000.

Economical inequality has an impact on, or even causes, the inequalities in housing conditions, health care and education. This paper is addressed the sector of health care.

### EVOLUTIONARY PROCEDURES IN ANALYSIS OF TIME SERIES WITH MISSING DATA

##### Anna Romowicz

Katowice University of Economics

Poland

The analysis of time series data is an important area of forecasting. But there is a common practical problem in time series analysis – that is missing data. There are many reasons of loosing information, from human mistakes till failure of monitoring devices. There are different missing-data methods, for example mean substitution, mean of adjacent observations, or maximum likelihood estimation. This paper suggest different approach to missing data problem: maybe it is possible to make satisfied forecast based on time series without filling up missing information? To answer this question some results of using evolutionary procedures will be presented.

### MODELLING OF VOLATILITY AT CZECH

FINANCIAL MARKETS

##### Jiri Tresl

Prague University of Economics

Czech Republic

At present, two principal approaches to volatility modelling exist: the GARCH and stochastic volatility models. Throughout this paper, the GARCH models have been applied both to selected Czech capital market and exchange rates time series. The first aim of investigation consists in possible differences in behaviour with respect to time scale (days, weeks, months).

The second level is given by differences between stock prices and exchange rates. Finally, besides univariate models, multivariate GARCH were also used to discover possible relations among different time series.

### DIFFERENTIATION OF HLE AND FACTORS INFLUENCING ON HLE

##### Paulina Ucieklak-Jeż

Jan Dlugosz University in Czestochowa

Poland

Healthy Life Expectancy is one of fundamental indicators of the population health state estimation. The aim of this article is to answer how HLE Healthy Life Expectancy depends on economic development. The increase of welfare imply the existence of good life conditions influencing on longer HLE estimation of studied population.

The health expenditure against HLE estimates were plotted. The relationship between them is consistent with the view that health expenditure has diminishing returns. It is positive up to on certain expenditure level per capita. Additional expenditure beyond this level has a negligible incremental effect on HLE.

### MINIMUM BIAS METHODS IN A’PRIORI RATE MAKING

##### Alicja Wolny-Dominiak

Katowice University of Economics

Poland

Insurance companies specialising in casualty insurance create their own rating systems for setting fair premiums for every risk for different kinds of insurance portfolios. The rating system is mostly based on the data analysis concerning the number and the value of claims for individuals or groups (classes) of insured people within a given portfolio. Based on a given rating system, the premium for a particular risk is calculated in two stages: a priori rating and a posteriori rating. In this paper, the process of a priori rating is analyzed with the emphasis on minimum bias methods used for modelling the rating variables.

### FOUNDATIONS OF STATISTICAL DESCRIPTION OF “AUSTRIAN” SILESIA

##### Prokop Závodský

Prague University of Economics

Czech Republic

In the late of XVIII. and early in XIX. century, extensive works dealing with statistical and topographic description of individual countries creating present-day Czech Republic were published. The part of Silesia left after Austrian-Prussian wars the constituent of Habsburg monarchy was the topic of four-volume work Topographie des k. k. Antheils von Schlesien written by piarist Reginald Kneifel (1761-1826).

### APPLICATION THE IRT MODELS WITH EXTERNAL AND ITERNAL FACTORS TO STATISTICAL INFERENCE WITH MISSING DATA

##### Beata Zmyślona

Wroclaw University of Economics

Poland

Definition of missing data mechanism is main problem in statistical inference with missing data. Every method of incomplete data analysis assumes one of two kinds of missing data mechanism implicitly or explicitly, namely random or non random mechanism. Assumptions about missing data mechanism depend on the reasons for an occurrence of missing data. In social-economical research this mechanism is generally non random and it should be taken into account in statistical inference. Statistical inference with missing data mechanism improves properties of estimators calculated on the base of incomplete data and allows to take into account the information about non-response reasons.

This paper presents the examples of application of the item response theory to model the missing data mechanism. In this theory we use latent variables to model the response propensity. We consider two groups of models, namely models with external factors and models with internal factors. Models with external factors enable taking into account the different types of variation, namely random and variation due to some fixed effects. Using external factors to explain variation requires the knowledge about what might play a role in the variation. When this knowledge is not available we can use the second group of models, namely models with internal factors.