C58 - Financial EconometricsReturn

In this article I have analysed the long-term relationship between US M2 money supply and the price of gold per troy ounce using Engle-Granger cointegration. The analysis shows the existence of long-term price dependency of gold in relation to US M2 money supply. M2 was used in two variants, seasonally adjusted and not seasonally adjusted. No relevant difference was observed between them. A period spanning 53 years, from 1970 to 2023, was analysed. An EC model using monthly observations indicates very weak correlation between the change of M2 and the subsequent change of the gold price, so semiannual observations were used instead which proved fully conclusive. This, together with results from the long-term model, confirms long price cycles and fluctuation around its equilibrium price lasting for years, which allows the use of gold as a hedge against increasing M2. This article may prove beneficial in filling the gap as it confirms gold price to be dependent on US M2 on longer time span analysed than previous studies and is fully able to explain gold price analysing dependency of two variables only.

Green finance is becoming increasingly important today, affecting many areas of the economy. In this regard, the examination of green bond markets is becoming more and more important, as various financial shocks have also led to significant changes in the financial markets and economic policy processes. However, only a few of these new financial assets were issued on the emerging EU market, therefore the side effects of them have not yet been fully explored. In addition, the rise of green finance is only in its infancy in smaller economies, in various financial markets, which may be important to monitor in future investment decisions. The aim of our study was to examine the volatility properties of green sovereign bonds of European small open economies for the period between 2016 and 2021, where we analysed how the differences of these green sovereign bonds to conventional sovereign bonds changed over time. Also, we wanted to test whether there was a possibility of a conditional volatility premium for green government bonds. To answer our research questions, we calculated conditional volatilities, and the green premiums towards their standard forms using GARCH models. Our result suggested that the Polish and Hungarian green sovereign bonds have higher volatility than the traditional ones, which is the opposite of the German experience.

Insurance and reinsurance markets are exposed to influence of unsystematic catastrophe risk, which is caused by adverse development in natural disasters. Deepening of influence is highlighted in relation to irregular and extreme impacts of catastrophes. Insurers face to coverage of random loss events with contribution of reinsurers through a transfer of specified part of non-life risk. Influence of catastrophe risk is assessed through analysis of annual and quarterly development in catastrophe losses, impacts of floods, amount of premium written and reinsurance premium with respect to relations among selected variables. Analysis is focused on specific conditions of the Czech insurance market in context of reinsurance market. The findings concern to negative development in increasing catastrophe and flood losses, which seems to be not in line with changes in premium written, respectively reinsurance premium, which remain at stable level during analysed period. Respecting unfavourable aspect of insufficient data, results emphasize inadequate level of reinsurance of significant catastrophe risk, respectively floods, within the Czech insurance market.

Echo State Neural Networks (ESN) were applied to forecast the realized variance time series of 19 major stock market indices. Symmetric ESN and asymmetric AESN models were constructed and compared with the benchmark realized variance models HAR and AHAR that approximate the long memory of the realized variance process with a heterogeneous auto-regression. The results show that asymmetric models generally outperform symmetric ones, indicating that a correlation between volatility and returns plays an important role for volatility forecasting. Additionally, models utilizing a logarithmic transformation of the time series achieved generally better results than models applied directly to the realized variance. Echo State Neural Networks outperformed HAR and AHAR models for several important indices (S&P500, DJIA and Nikkei indices), but on average they achieved slightly worse results than the AHAR model. Nevertheless, the results show that Echo State Neural Networks represent an easy-to-use and accurate tool for realized variance forecasting, whose performance may potentially be further improved with meta-parameter optimization.

The aim of this paper is to analyze the structural determinants of the total loans volume in the Czech banking sector. Analysis of five selected characteristics is realized in period 2000-2008. It used the OLS regression analysis for estimate of model. The regression analysis showed that the concentration of the credit market and the profitability calculated as the return on assets (ROA) has positive impact to total loans and the quality of portfolio has negative impact to total loans. If the share of classified loans to total loans is decreasing, total loans volume is rising and if concentration ratio of the credit market and return of assets are rising, total loans volume is also rising.