Statistika: Statistics and Economy Journal - No. 4/2024

Statistika, 104(4): 383–398
https://doi.org/10.54694/stat.2024.41

Abstract
In the period following 2000, the Czech economy went through two crises, which were different in their causes, durations and consequences. The 2009–2013 crisis resulted from the global financial and fiscal crisis. Its causes were external and purely economic, and its impact on households' economic behaviour was 'standard' – a gradual and moderate reduction in consumption and investment with high unemployment and low inflation rates. The advent of the COVID-19 pandemic in 2020 meant a sudden and unexpected change in economic conditions – the closure of shops and services, restricted population movements, and household consumption limited to only the most essential products. A reduction in household consumption generally means an increase in household savings if all other circumstances are equal. The aim of the present paper is to show, using the methods of time series analysis, the effects of these two crisis periods in terms of data for the household sector or to show whether the fall in the propensity to consume and the rise in the propensity to save in 2020–2021 can be considered statistically significant compared to the crisis period of 2009–2013. Publicly available data from the Czech Statistical Office have been used for our analysis.

Keywords
National accounts, households, final consumption expenditure, gross saving, time series analysis, time series stability tests

Statistika, 104(4): 399–409
https://doi.org/10.54694/stat.2024.53

Abstract
The Covid-19 pandemic triggered a massive economic policy response in all developed countries. There were various approaches that governments adopted with a variety of possible aims and outcomes. In our research we focus on estimating the effects of Covid-19 on the Czech Republic, a small open economy, which was typical in that it had high mortality, a long lockdown and a focus on job preservation. Using three different methods – input-output, CGE and ADL, we estimate the industry-level impact of the pandemic, taking induced effects into account. We show that, besides industries that were explicitly harmed like accommodation or transportation, some industries like construction seem to be implicitly vulnerable as well. This is an important finding especially for any future policy responses. Regarding the economic policy itself, we conclude that it was successful in terms of preserving jobs, but the expenditures were probably too high to call it an efficient policy response.

Keywords
Labor market, economic policy, Covid pandemic, CGE model, econometrics

Statistika, 104(4): 410–423
https://doi.org/10.54694/stat.2024.11

Abstract
This paper contributes to the recent empirical literature on the absorption determinants of the European Cohesion Policy (ECP). In particular, it attempts to verify the effect of the digitalization of the public sector in the form of e-government services on the ECP absorption rates during the period 2007–2016. Using the fixed effects panel models, we confirm that increased usage of e-government services is associated with higher absorption rates of the beneficiaries. Moreover, we reveal its conditional effects in connection with government quality, human capital, and recessionary periods. While its positive effect is neutralized in the recession, the benefits of the e-government use towards ECP absorption are more pronounced at a higher level of government quality and a share of skilled labor. The results therefore suggest that the promotion of digitalization and training can not only promote economic growth and innovation as commonly known but, as the analysis shows, can also be valuable in terms of ECP absorption rates.

Keywords

Digitalization, European Cohesion Policy, e-government, absorption

Statistika, 104(4): 424–439
https://doi.org/10.54694/stat.2024.22

Abstract
Different reserving methods can be used to predict claim values in non-life insurance. This article compares two different methodological approaches to reserving methods, namely, Chain-ladder (the traditional approach to reserving in non-life insurance) and state-space modeling (the modern approach based on recursive Kalman filtering). Moreover, the paper compares both methods with the involvement of clustering which divides claims into several groups according to their similarity and ensures greater homogeneity of data. To be able to compare the accuracy of reserve predictions numerically one suggests three types of generators of large insurance portfolios that represent well the behavior of the given methods in practice (one of them is derived directly from a real Czech non-life insurance claims portfolio). The obtained results may serve as a hint to improve the state-space methodology in order to give comparable results with classical approaches to reserving since in future the state-space modeling will be important for micro reserving where the “clustering” gains nearly a form of individual policy contracts.

Keywords
Chain-ladder, claims portfolio generators, clustering, loss reserving, non-life insurance, state-space model

Statistika, 104(4): 440–464
https://doi.org/10.54694/stat.2024.30

Abstract
Research background: Bankruptcy and financial distress prediction has always been an integral part of any financial management system. It gives an indication to stakeholders to take precautionary measures in order to avoid losses. The traditional approaches for prediction, including logistic regression and discriminant analysis, are constrained by their inability to deal with complex and high-dimensional data (Odom and Sharda, 1990; Min and Lee, 2005). Recent developments in the field of machine learning, and particularly autonomous learning classifiers, present a potential proposed alternative.
Purpose: The purpose of this paper is to propose a first-order autonomous learning classifier (F-O ALMM0) for predicting bankruptcy of business entities and individuals.
Design/methodology/approach: The data file contained a total of 352 companies obtained from the Kaggle database and incorporating 83 financial ratios. Initially, the model's performance was assessed as a preliminary step, but the results were average, followed by the application of Principal Component Analysis (PCA) to enhance the quality of the input’s variables. Afterwards, the number of independent variables was reduced to 26. Thus, the results were improved.

Keywords
Bankruptcy prediction, first-order, autonomous learning, Multi-Model Classifier, Principal Components Analysis

Statistika, 104(4): 465–479
https://doi.org/10.54694/stat.2024.10

Abstract
This article explores the impact of trend inflation on monetary policy under a higher inflation target. Adding trend inflation to DSGE models helps us understand the inflation gap better; the gap is less persistent when it is measured as a deviation from trend instead of as a constant average. A high inflation target is likely to overshoot unless the monetary authorities adopt restrictive measures to keep output below its deterministic equilibrium. Indeed, Bank Al-Maghrib raised its key rate by 0.25 percentage points to achieve an inflation rate of 2%, underscoring the importance of maintaining this trajectory. The study identifies key policy implications: higher trend inflation destabilizes expectations, forcing monetary policy to react more to inflation deviations and less to output gaps in high-target environments. These conclusions hold for different parameterizations and specifications of the Taylor rule (backward-looking, forward-looking, and inertial). In addition, Taylor rules based on output growth rather than output gaps widen the zone of determinism, making it easier to adopt a single reference value.

Keywords
Trend inflation, monetary policy analysis, economic stability assessment, Morocco's DSGE model

Statistika, 104(4): 480–495
https://doi.org/10.54694/stat.2023.58

Abstract
This paper examines the access to higher education across socio-religious groups in the state of Jharkhand in India. It also examines the factors affecting access to higher education and the role of students’ social background in explaining the inequality in participation in higher education. The analysis is based on cross tabulation, logistic regression and Fairlie decomposition method. The analysis shows that tribals, Muslims and Scheduled Castes are the worst performing groups in the state. The most prominent factor behind the vulnerable condition of tribals is their high concentration in rural areas as there is a remarkable gap in their performance between rural and urban areas. A large part of the gap between the privileged and the underprivileged groups could not be explained by endowment factors, namely, household size, education of the head of household and income background. The results suggest that incentives created due to family background leads to different outcomes among different socio-religious groups.

Keywords
Higher education, human capital, inequality, discrimination, Logistic Regression Model, Fairlie decomposition method

Statistika, 104(4): 496–502
https://doi.org/10.54694/stat.2024.35

Abstract
The year 2024 marks twenty years since Czechia joined the EU. This anniversary offers an opportunity to evaluate the benefits and losses associated with the membership, and a sufficiently long time series allows such an evaluation to be made on representative data sets. Among others, the evaluation of the membership based on drawing financial funds from the EU budget stands out. Since joining the EU in 2004, the Czech Republic has received one trillion crowns more from the EU budget than it has paid into it. At the same time, in no year since 2004 the balance of income and expenditure in relation to the EU budget has been negative. This article describes basic principles of the EU funds and deals with accrual recording in national accounts and the cash flow concept. In the context of 20 years of EU membership, it then evaluates the net position of Czechia in relation to the EU and the influence of income from the EU on the revenue of general government.

Keywords
European Union, general government, national accounts, EU funds