|
|
Forecasting realized volatility using machine learning and mixed-frequency data (the case of the Russian stock market)
Pyrlik, Vladimir ; Elizarov, P. ; Leonova, A.
We assess the performance of selected machine learning algorithms (lasso, random forest, gradient boosting, and long short-term memory) in forecasting the daily realized volatility of returns of selected top stocks in the Russian stock market in comparison with a heterogeneous autoregressive realized volatility benchmark in 2018-2020. We seek to improve the predictive power of the models by including various economic indicators that carry information about future volatility. We find that lasso delivers a good combination of easy implementation and forecast precision. The other algorithms require fine-tuning and frequent re-training, otherwise they are likely to fail to outperform the benchmark often enough. Only the basic lagged log-RV values are significant explanatory variables in terms of the benchmark in-sample quality. Many economic indicators of mixed frequencies improve the predictive power of lasso though, including calendar and overnight effects, financial spillovers from local and global markets, and various macroeconomics indicators.
|
|
|
Analysis of Interdependencies among Central European Stock Markets
Mašková, Jana ; Baruník, Jozef (advisor) ; Princ, Michael (referee)
The objective of the thesis is to examine interdependencies among the stock markets of the Czech Republic, Hungary, Poland and Germany in the period 2008-2010. Two main methods are applied in the analysis. The first method is based on the use of high-frequency data and consists in the computation of realized correlations, which are then modeled using the heterogeneous autoregressive (HAR) model. In addition, we employ realized bipower correlations, which should be robust to the presence of jumps in prices. The second method involves modeling of correlations by means of the Dynamic Conditional Correlation GARCH (DCC-GARCH) model, which is applied to daily data. The results indicate that when high-frequency data are used, the correlations are biased towards zero (the so-called "Epps effect"). We also find quite significant differences between the dynamics of the correlations from the DCC-GARCH models and those of the realized correlations. Finally, we show that accuracy of the forecasts of correlations can be improved by combining results obtained from different models (HAR models for realized correlations, HAR models for realized bipower correlations, DCC-GARCH models).
|
|
|
Forecasting realized volatility: Do jumps in prices matter?
Lipták, Štefan ; Baruník, Jozef (advisor) ; Šopov, Boril (referee)
This thesis uses Heterogeneous Autoregressive models of Realized Volatility on five-minute data of three of the most liquid financial assets - S&P 500 Futures index, Euro FX and Light Crude NYMEX. The main contribution lies in the length of the datasets which span the time period of 25 years (13 years in case of Euro FX). Our aim is to show that decomposing realized variance into continuous and jump components improves the predicatability of RV also on extremely long high frequency datasets. The main goal is to investigate the dynamics of the HAR model parameters in time. Also, we examine whether volatilities of various assets behave differently. Results reveal that decomposing RV into its components indeed improves the modeling and forecasting of volatility on all datasets. However, we found that forecasts are best when based on short, 1-2 years, pre-forecast periods due to high dynamics of HAR model's parameters in time. This dynamics is revealed also in a year-by-year estimation on all datasets. Consequently, we consider HAR models to be inappropriate for modeling RV on such long datasets as they are not able to capture the dynamics of RV. This was indi- cated on all three datasets, thus, we conclude that volatility behaves similarly for different types of assets with similar liquidity. 1
|
|
|
The impact of renewable resources on price volatility in the European power markets
Líšková, Katarína ; Krištoufek, Ladislav (advisor) ; Luňáčková, Petra (referee)
Integration of renewable energy sources impacts electricity spot price and its variation. Remaining open question is, in which direction. Volatility fluctuations threaten secur- ity of electricity supply, influence trading strategies and create uncertainty in optimal installed capacity planning. In this thesis, drivers of price volatility in Czech and Ger- man day-ahead power market are analysed with an emphasis on penetration of renewable energy sources. To the best of our knowledge, this is the first study focused on this issue in Czech electricity market. We apply recently developed approach of quadratic variation theory with an adjustment for electricity prices. Realised volatility is divided into its continuous and jump component. The continuous part is modelled by three het- erogeneous autoregressive models, differing in complexity and inclusion of market-specific fundamental variables. Amendments to each model for the particular market are proposed and the models are evaluated both in-sample and out-of-sample. Addition of exogenous variables − commodity prices, weather conditions and seasonal variables − to simpler heterogeneous autoregressive model is found to improve volatility forecast accuracy. The results suggest higher continuous volatility due to increased penetration of power from wind...
|
|
|
Forecasting realized volatility: Do jumps in prices matter?
Lipták, Štefan ; Baruník, Jozef (advisor) ; Šopov, Boril (referee)
This thesis uses Heterogeneous Autoregressive models of Realized Volatility on five-minute data of three of the most liquid financial assets - S&P 500 Futures index, Euro FX and Light Crude NYMEX. The main contribution lies in the length of the datasets which span the time period of 25 years (13 years in case of Euro FX). Our aim is to show that decomposing realized variance into continuous and jump components improves the predicatability of RV also on extremely long high frequency datasets. The main goal is to investigate the dynamics of the HAR model parameters in time. Also, we examine whether volatilities of various assets behave differently. Results reveal that decomposing RV into its components indeed improves the modeling and forecasting of volatility on all datasets. However, we found that forecasts are best when based on short, 1-2 years, pre-forecast periods due to high dynamics of HAR model's parameters in time. This dynamics is revealed also in a year-by-year estimation on all datasets. Consequently, we consider HAR models to be inappropriate for modeling RV on such long datasets as they are not able to capture the dynamics of RV. This was indi- cated on all three datasets, thus, we conclude that volatility behaves similarly for different types of assets with similar liquidity. 1
|
|
|
Analysis of Interdependencies among Central European Stock Markets
Mašková, Jana ; Baruník, Jozef (advisor) ; Princ, Michael (referee)
The objective of the thesis is to examine interdependencies among the stock markets of the Czech Republic, Hungary, Poland and Germany in the period 2008-2010. Two main methods are applied in the analysis. The first method is based on the use of high-frequency data and consists in the computation of realized correlations, which are then modeled using the heterogeneous autoregressive (HAR) model. In addition, we employ realized bipower correlations, which should be robust to the presence of jumps in prices. The second method involves modeling of correlations by means of the Dynamic Conditional Correlation GARCH (DCC-GARCH) model, which is applied to daily data. The results indicate that when high-frequency data are used, the correlations are biased towards zero (the so-called "Epps effect"). We also find quite significant differences between the dynamics of the correlations from the DCC-GARCH models and those of the realized correlations. Finally, we show that accuracy of the forecasts of correlations can be improved by combining results obtained from different models (HAR models for realized correlations, HAR models for realized bipower correlations, DCC-GARCH models).
|
|
|
Forecasting realized volatility: Do jumps in prices matter?
Lipták, Štefan ; Baruník, Jozef (advisor) ; Šopov, Boril (referee)
This thesis uses Heterogeneous Autoregressive models of Realized Volatility on five-minute data of three of the most liquid financial assets - S&P 500 Futures index, Euro FX and Light Crude NYMEX. The main contribution lies in the length of the datasets which span the time period of 25 years (13 years in case of Euro FX). Our aim is to show that decomposing realized variance into continuous and jump components improves the predicatability of RV also on extremely long high frequency datasets. The main goal is to investigate the dynamics of the HAR model parameters in time. Also, we examine if volatilities of various assets behave differently. The results reveal that decomposing RV into its components indeed im- proves the modeling and forecasting of volatility on all datasets. However, we found that forecasts are best when based on short, 1-2 years, pre-forecast periods due to high dynamics of HAR model's parameters in time. This dynamics is revealed also by a year-by-year estimation on all datasets. Con- sequently, we consider HAR models to be inapproppriate for modeling RV on such long datasets as they are not able to capture the dynamics of RV. This was indicated on all three datasets, thus, we conclude that volatility behaves similarly for different types of assets with similar liquidity. 1
|
|
|
Analysis of Interdependencies among Central European Stock Markets
Mašková, Jana ; Baruník, Jozef (advisor) ; Princ, Michael (referee)
The objective of the thesis is to examine interdependencies among the stock markets of the Czech Republic, Hungary, Poland and Germany in the period 2008-2010. Two main methods are applied in the analysis. The first method is based on the use of high-frequency data and consists in the computation of realized correlations, which are then modeled using the heterogeneous autoregressive (HAR) model. In addition, we employ realized bipower correlations, which should be robust to the presence of jumps in prices. The second method involves modeling of correlations by means of the Dynamic Conditional Correlation GARCH (DCC-GARCH) model, which is applied to daily data. The results indicate that when high-frequency data are used, the correlations are biased towards zero (the so-called "Epps effect"). We also find quite significant differences between the dynamics of the correlations from the DCC-GARCH models and those of the realized correlations. Finally, we show that accuracy of the forecasts of correlations can be improved by combining results obtained from different models (HAR models for realized correlations, HAR models for realized bipower correlations, DCC-GARCH models).
|
guest ::
