C11 - Bayesian Analysis: GeneralReturn

In the last two decades, advances in globalization have evolved remarkably and countries have become more integrated with the entire world. The implications of this process have attracted interest of researchers and monetary policy authorities. This paper provides an assessment of the impact of globalization on macroeconomic dynamics at the level of four representative Central Eastern European countries (Czechia, Hungary, Poland and Romania) between 2003Q2 and 2022Q4. The method proposed in this study is a time-varying vector autoregressive model with stochastic volatility estimated using the Metropolis-Hastings algorithm through Bayesian inference. The results of the impulse response functions suggest a slight decrease in inflation and an increase in economic activity for some of the analysed countries, after a positive shock in the degree of trade openness, a proxy used for the advance in globalization.

The main objective of this study is to determine the effects of economic and social inequalities on academic success and to test whether the cycle of inequality is active through education. This objective is accomplished using classical and Bayesian discrete choice models for the sample obtained from Türkiye. The results reveal that students' economic and social characteristics affect their academic success and that these characteristics are possible sources of inequality in education. According to the findings obtained from models employed in the study, income, private school education, parental education level, region of residence, neediness to work, and the level of happiness with the family were found to have statistically significant effects on student success in getting into the desired university department and university placement ranking. Additionally, the results are compatible with the studies that report that the Bayesian approach yields more stable and appropriate results with smaller standard errors and confidence intervals.

Using the factor-augmented vector autoregressive (FAVAR) model proposed by Bernanke et al. (2005), this study explores the effect of monetary policy on a wide range of macroeconomic and financial variables for the US, Canada and the UK. The study makes use of financial data from 1990 to 2016, comprising 55-70 variables of the three major nations to show (1) that factors come with additional informational capability, which summarizes the performance of key macroeconomic variables and (2) the manner in which these variables are affected by contractionary monetary policies. Our findings confirm that monetary policy tightening results in decrease in industrial production, employment, share prices, housing starts and inflation; however, it leads to increase in the three-month treasury bill rate, long-term interest rates and unemployment. Overall, the impact of standardized monetary tightening is similar across the countries studied. These results from the major economies and the inclusion of larger data sets containing more variables would be relevant for policy theorists and practitioners from other countries.

This paper analyses the unemployment hysteresis in the Czech Republic on data from 1999 to 2016. The hysteresis is modelled by allowing for the impact of cyclical unemployment on the non-accelerating inflation rate of unemployment. Models are estimated using the Bayesian approach and provide robust evidence in favour of the hysteresis. The estimates imply that in response to an increase in the cyclical unemployment of 1 percentage point, the non-accelerating inflation rate of unemployment increases by 0.18 percentage points.

This paper investigates the impact of behavioural attention on the households' foreign currency savings as a response to the external macroeconomic shocks. The information that the households acquire via different communication channels is expected to influence their decisions regarding their savings' allocation into different currencies. This study has applied the fundamental macroeconomic models by including individuals' attention to the specific risks and search interest in specific keywords on Google in order to assess the impact of acquired information and its communication channel on the households' foreign currency savings. We employed a two-level mixed effects model including macroeconomic fundamentals and individuals' attention to the information determinants. We solved a problem of a long list of potential explanatory variables (keywords) by employing the Bayesian Model Averaging. This study assumes that households are more sensitive to the macroeconomic shocks (factors) if they search simultaneously for infor-mation on Google about these factors or specific related risks. The results emphasize the role behavioural attention during financial turmoil and economic downturn periods, especially in the environment of very low interest rates.

This article identifies structural changes in the Czech economy in the period from 1996 to 2012 using a DSGE model estimated using Bayesian methods. A structural change is understood as a statistically significant change in model parameter(s). Prior to the first quarter of 1999, there was a structural change that can be primarily attributed to shocks impacting only the domestic economy, and to the domestic monetary authority's increased preferences towards inflation and exchange rate growth. The elasticity of substitution between domestic and imported consumption goods also increased sharply in this period. As far as the recent economic recession is concerned, it was caused by a much more persistent worldwide technology shock. Habit formation dropped abruptly during the crisis as households tended not to smooth their consumption much anymore. Recursive impulse response analysis carried out on the model suggests that the propagation mechanisms in the model economy changed, implying that the identified structural changes were accompanied by a change in behaviour of the model economy.

A new good quality standardized data set for cross-national comparisons of income inequality covering 147 countries has been constructed. The Bayesian model averaging and multiple imputation approach have been used to identify robust correlates from a larger pool of potential predictors of cross-national variation in inequality determined from previous literature. The results document signiicant macro-regional speciicity both regarding levels and predictors of inequality. While globalization associates with lower inequality in Western countries, it has opposite e ects in Latin America, Sub-Saharan Africa or post-communist countries. Age structure, the extent of social spending, or colonial history are another important factors with regionally speciic impacts on inequality. By contrast, explanative power of some other traditional determinants of inequality has not been corroborated.

We formulate a bivariate stochastic volatility jump-diffusion model with correlated jumps and volatilities. An MCMC Metropolis-Hastings sampling algorithm is proposed to estimate the model's parameters and latent state variables (jumps and stochastic volatilities) given observed returns. The methodology is successfully tested on several artificially generated bivariate time series and then on the two most important Czech domestic financial market time series of the FX (CZK/EUR) and stock (PX index) returns. Four bivariate models with and without jumps and/or stochastic volatility are compared using the deviance information criterion (DIC) confirming importance of incorporation of jumps and stochastic volatility into the model.

EVA is a useful tool for assessing company performance. It combines factors, such as economy, accounting and market information in its assessment. This study employed EVA in an attempt to compare the companies' performances of GLCs (government-linked companies) and non-GLCs. Based on a 4-year pooled panel data of 37 GLCs and 208 non-GLCs, the results show that companies with government as their stakeholders tend to exhibit lower EVA scores than the companies without government stakeholders in Malaysia. Larger size companies were found to have lower EVA values. Companies which have both the characteristics - which are simultaneously large in size and government-owned, tend to be most adversely affected. Thus, any increment in the size of company for GLCs would decrease or destroy the value of the company, and to a greater degree, than companies without government holding.