Statistika: Statistics and Economy Journal - No. 1/2017

The paper deals with statistical data on regional employment that was constructed on the basis of regional input-output tables. Both regional input-output tables and product linked regional employment were constructed within the research project. This data fits well the purposes of detailed analysis of regional economy since the data is broken down by two-digits level of product classification (CZ-CPA). Employment is presented on the level of the regions (NUTS 2) of the Czech Republic for 2011. The paper briefly describes procedures allowing construction of regional employment by products and the links to data from official statistics. Some of analytical possibilities of data on regional employment are illustrated by simple input-output analysis with three scenarios. The regions are also tested for output and employment sensitivity by estimating multipliers and elasticities. The interpretation of obtained results including hierarchical clustering is provided. The paper also presents discussion about the use of regional input-output tables and regional employment in regional analyses for policy measures.

Keywords
Regional input-output tables, employment, input-output analysis

Analytical works usually use single-regional approach which does not demand so much data. However, this approach disregards flows of output among regions. This leads to a misrepresentation of results which can be eliminated by using Inter-regional input-output model that requires more data to be employed. This paper illustrates the differences between the two different approaches of regional input-output model construction and their results. We construct inter-regional and single-regional models for all 14 regions of the Czech Republic and with 82 products according to the Classification of Products CZ-CPA. The results are compared on the level of Leontief’s matrix and multipliers. We use graphical illustrations to depict the systematicness of differences. The single-regional approach proves a systematic undervaluation of specific products and regions contrary to other regions. The graphical analysis shows the significance of the connection among regions. This illustrates the disadvantage of the single regional approach. Finally, the results confirm the idea of a signifiant analytical misrepresentation of impacts modelled by this approach in the case of data for the Czech Republic.

Keywords
Regional Input-Output Tables, Input-Output analysis, Leontief`s multipliers, IRIO

The main goal of this paper is to present an analysis of financial quarterly time series describing the level of book leverage of U.S. companies selected from different industries in the period 1991–2014. The basic question is whether the sub-prime crisis 2007–2008 caused a change in the behavior of the respective companies. More generally, we are interested whether the time series may be considered stationary. Statistical methods suitable for the detection of breaks (changes) for individual and panel data are presented together with their pros and cons. Against our expectations, the analysis did not reveal a significant change due to the sub-prime crisis. On the other hand, all series contain at least one change, most of the changes occurring around the year 2000, thus offering room for an economic explanation.

Keywords
Change point problem; abrupt, gradual and multiple changes; stationarity in the mean; sum and maximal test statistics; panel data; book leverage

To overcome drawbacks of central moments and comoment matrices usually used to characterize univariate and multivariate distributions, respectively, their generalization, termed L-moments, has been proposed. L-moments of all orders are defined for any random variable or vector with finite mean. L-moments have been widely employed in the past 20 years in statistical inference. The aim of the paper is to present the review of the theory of L-moments and to illustrate their application in parameter estimating and hypothesis testing. The problem of estimating the three-parameter generalized Pareto distribution’s (GPD) parameters that is generally used in modelling extreme events is considered. A small simulation study is performed to show the superiority of the L-moment method in some cases. Because nowadays L-moments are often employed in estimating extreme events by regional approaches, the focus is on the key assumption of index-flood based regional frequency analysis (RFA), that is homogeneity testing. The benefits of the nonparametric L-moment homogeneity test are implemented on extreme meteorological events observed in the Czech Republic.

Keywords
L-moment, parameter estimation, generalized Pareto distribution, homogeneity testing, precipitation extreme events, Czech Republic

One of the approaches for forecasting future values of a time series or unknown spatial data is kriging. The main objective of the paper is to introduce a general scheme of kriging in forecasting econometric time series using a family of linear regression time series models (shortly named as FDSLRM) which apply regression not only to a trend but also to a random component of the observed time series. Simultaneously performing a Monte Carlo simulation study with a real electricity consumption dataset in the R computational langure and environment, we investigate the well-known problem of “negative” estimates of variance components when kriging predictions fail. Our following theoretical analysis, including also the modern apparatus of advanced multivariate statistics, gives us the formulation and proof of a general theorem about the explicit form of moments (up to sixth order) for a Gaussian time series observation. This result provides a basis for further theoretical and computational research in the kriging methodology development.

Keywords
Forecasting models, linear regression models, best linear unbiased prediction, approximation of mean squared error, moments of random vectors

Nowadays, the air transportation is one of key means of transport. Unfortunately, there are many factors influencing its popularity and intensity of its use. There are many studies investigating these factors. The paper investigates the possibility of classifying the world's airports in terms of the trend in the number of handled passengers as it is one of the main economic indicators for airports. For this classification we chose cluster analysis. The paper focuses on an important aspect of the process, which chooses the appropriate number of clusters. It turned out that in terms of interpretation of the results, it may not always be the most efficient to set this number at the theoretically best and recommended value. As a result of our analysis, several groups of airports with similar trend of post-event reactions are found. Therefore, this may bring better understanding of the intensity and the range of the impact of particular events on air transportation.

Keywords
Cluster analysis, number of clusters, occupancy of airports, Bayesian Information Criterion, Akaike Information Criterion, silhouette coefficient

This study contributes to the empirical literature on augmented neo-classical production function. It is done by introducing steel production into macro-production function of the European Union. The data is collected from the World Development Indicators and the World Steel Association from the period of 1980–2014. We apply second generations of unit root tests to examine stationarity and panel cointegration with cross-sectional dependence to analyze long run relationship between national income and steel production. Robustness of tests is also reached by using 23 estimators and country specific slopes. Whereas, to detect the cause and effect, Granger and Dumitrescu-Hurlin causality tests are applied. Uni-directional causality from national income to steel production is found. Recommendations are made on the basis of empirical results.

Keywords
Steel production, national income, augmented mean group, panel causality

The economic returns to schooling is a fundamental parameter of interest in many different areas of economics and public policy. The most common technique for estimating this parameter is based on the assumption that the ‘true’ coefficient of education in the earnings equation is constant across individuals. However, this may not often be wholly true and returns to schooling estimates may be biased and inconsistent. The objective of this study was to estimate the returns to schooling as a random coefficient and obtain accurate and reliable estimates that will be useful for policy recommendations. The restricted maximum likelihood (REML) method was used to estimate the parameters of a random coefficient model using data from a 2007/2008 Ghanaian twins’ survey. The results revealed that the REML economic returns to schooling in three selected cities were between 7% and 9%. Significant (p<0.05) variances around the mean returns to schooling implied that returns to schooling might vary among individuals due to unobserved factors.

Keywords
Return to schooling, Random coefficient model, Maximum Likelihood, REML, Variance Heterogeneity