National Repository of Grey Literature 105 records found  1 - 10nextend  jump to record: Search took 0.00 seconds. 
Application of machine learning methods for estimating apartment prices in the Czech Republic
Nikodym, Jakub ; Krištoufek, Ladislav (advisor) ; Baruník, Jozef (referee)
In this thesis, we propose alternative ways to apartments' mass appraisal. This work enriches the current literature by combining several techniques of data extraction and price estimation. We are not aware of any similar work providing an in-depth overview of the Czech apartment market. Throughout the empirical analysis, five different methods (OLS, LASSO, decision tree, random forests, and kNN) are applied to the dataset of 15,848 classifieds. The aim of the study is to find the most accurate method of esti- mating offering prices, using structured variables as well as data extracted by text mining. We use various accuracy statistics and graphical analysis to vali- date our results. Tree-based methods, specifically the random forest algorithm, results with the highest accuracy in predicting offering prices. Additionally, text-based variables included in the model cause the reduction of errors on linear models. The last part of the analysis covers the main determinants of property value in Prague and the rest of the Czech Republic. We show that prices in Prague can be estimated with higher preciseness and with the lower number of independent variables.
Can Model Combination Improve Volatility Forecasting?
Tyuleubekov, Sabyrzhan ; Baruník, Jozef (advisor) ; Červinka, Michal (referee)
Nowadays, there is a wide range of forecasting methods and forecasters encounter several challenges during selection of an optimal method for volatility forecasting. In order to make use of wide selection of forecasts, this thesis tests multiple forecast combination methods. Notwithstanding, there exists a plethora of forecast combination literature, combination of traditional methods with machine learning methods is relatively rare. We implement the following combination techniques: (1) simple mean forecast combination, (2) OLS combination, (3) ARIMA on OLS combined fit, (4) NNAR on OLS combined fit and (5) KNN regression on OLS combined fit. To our best knowledge, the latter two combination techniques are not yet researched in academic literature. Additionally, this thesis should help a forecaster with three choice complication causes: (1) choice of volatility proxy, (2) choice of forecast accuracy measure and (3) choice of training sample length. We found that squared and absolute return volatility proxies are much less efficient than Parkinson and Garman-Klass volatility proxies. Likewise, we show that forecast accuracy measure (RMSE, MAE or MAPE) influences optimal forecasts ranking. Finally, we found that though forecast quality does not depend on training sample length, we see that forecast...
On the Utilization of Machine Learning in Asset Return Prediction on Limited Datasets
Petrásek, Lukáš ; Baruník, Jozef (advisor) ; Novák, Jiří (referee)
In this thesis, we conduct a comparative analysis of how various modern ma- chine learning techniques perform when employed to asset return prediction on a relatively small sample. We consider a broad selection of machine learn- ing methods, including e.g. elastic nets, random forests or recently highly popularized neural networks. We find that these methods fail to outperform a simple linear model containing only 5 factors and estimated via ordinary least squares. Our conclusion is that applications of machine learning in fi- nance should be conducted carefully, because the techniques may not actually be as powerful as one might think when they are applied under unfavorable circumstances. JEL Classification C45, C52, C53, C58, G12 Keywords asset pricing, machine learning, return predic- tion, regression, decision tree, random forest, neural network Title On the Utilization of Machine Learning in Asset Return Prediction on Limited Datasets Author's e-mail Supervisor's e-mail
The Impact of German Renewable Electricity on Czech Electricity Spot Prices
Kouřílek, Matěj ; Baruník, Jozef (advisor) ; Janda, Karel (referee)
Thesis investigates the impact of German wind and solar energy on the dynamics of Czech electricity spot prices in the period between 2015 to 2018. Using pooled panel-GARCH model, a negative merit order effect of German wind and solar energy were observed. More specifically, one additional GW of power produced by wind and solar, reduces the spot price by 0.60 and 0.45 EUR/MWh, respectively. The negative merit order effect was also found in the case of Czech solar energy. Corresponding spot price reduction equals to 1.42 EUR/MWh per additional gigawatt hour. Next, increased volatility in the spot prices was found due to both German wind and Czech solar energy. I also observed that these effects differ during a day. Furthermore, I estimated total financial impact stemming from the negative merit order effect and compared it with the total costs of households that arise in surcharges to support renewable energy. While Czech households pay approximately 270 million euros annually in surcharges, the total financial impact stemming from the merit order is around 145 million euros. The value comprises the merit order effect of both Czech and German renewable sources. In other words, Czech and German households bear the costs of subsidized renewable energy while they do not necessarily profit on the merit...
Are realized moments useful for stock market returns analysis?
Saktor, Ira ; Baruník, Jozef (advisor) ; Kočenda, Evžen (referee)
This thesis analyzes the use of realized moments in asset pricing. The analysis is done using dataset containing log-returns for 29 of the most traded stocks and covering 10 years of data. The dataset is split into training set covering 7 years and test set covering 3 years of data. For each of the stocks a separate time series model is estimated. In evaluation of the quality of the models, metrics such as RMSE, MAD, accuracy in forecasting the sign of future returns, and returns achievable by executing trades based on the recommendations from the model are used. Even though the inclusion of realized moments does not provide significant improvements in terms of RMSE, it is found that realized skewness and kurtosis significantly contribute to explaining the returns of individual stocks as they lead to consistent improvements in identifying future positive, as well as negative, returns. Moreover, the recommendations from the models using realized moments can help us achieve significantly higher returns from trading stocks. Inclusion of the interaction terms for variance and returns, skewness and returns, and kurtosis and variance, provides additional improvement of forecasting accuracy, as well as improvements in returns achievable by executing transactions based on recommendations from the model....
Using CAPM for assessment of efficiency of managed portfolios-mutual funds
Pergl, David ; Gapko, Petr (advisor) ; Baruník, Jozef (referee)
This bachelor thesis tested hypothesis if 30 randomly selected equity funds outperformed the market systematically in the time period 2003-2018. Funds were divided into two groups with respect to their investment strategies (Small caps and Large caps) and were tested in periods of Bull and Bear markets. As a theoretical concept the Capital Asset Pricing model (CAPM) was used. Two parameters of its equation were tested, alpha coefficient as an indicator of managers' skills and fund expenses and beta coefficient as an indicator of level of risk. The CAPM equation was expanded by dummy variables to measure the effects of different investment strategies and market conditions. The thesis used panel data analysis as an approach of estimation of the parameters with Fixed and Random Effects models. Funds invested mainly on the U.S. market. Their prices were transformed to fund returns as required by the CAPM model and compared with returns of S&P500. Statistically significant results confirmed that the CAPM fitted the expected relationship of market and fund returns. It showed that the funds taking higher risk were rewarded by higher expected returns expressed by beta greater than 1. It also showed that the managers invested more carefully in the periods of Bear market. Values of alphas revealed that Large...
Frequency connectedness and cross section of stock returns
Haas, Emma ; Baruník, Jozef (advisor) ; Kukačka, Jiří (referee)
The thesis presents a network model, where financial institutions form linkages at various investment horizons through their interdependence measured by volatility connectedness. Applying the novel framework of frequency connectedness mea- sures Baruník & Křehlík (2018), based on spectral representation of variance de- composition, we show fundamental properties of connectedness that originate in heterogeneous frequency responses to shocks. The newly proposed network mod- els characterize financial connections and systemic risk at the short-, medium- and long-term frequency. The empirical focus of this thesis is on the interde- pendence structure of US financial system, specifically, major U.S. banks in the period 2000 - 2016. In the light of frequency volatility connectedness measures, we argue that stocks with high levels of long-term connectedness represent greater systemic risk, because they are subject to persistent shocks transmitted for longer periods. When we assess institutions' risk premiums in asset pricing model, the model confirms the significance of volatility connectedness factor for asset prices. JEL Classification C18, C58, C58, G10, G15, Keywords connectedness, frequency, spectral analysis, sys- temic risk, financial network Author's e-mail Supervisor's e-mail...
Frequency Connectedness of Financial, Commodity, and Forex Markets
Šoleová, Juliána ; Baruník, Jozef (advisor) ; Vácha, Lukáš (referee)
This Thesis is dedicated to the variance decompositions from the VAR model un- der the Diebold, Yilmaz (2012) methodology combined with the Baruník, Křehlík (2017) method of frequencies that was used to create traditional and directional spillover tables to be compared under different frequencies. Diverse markets vari- ables were used for the analysis during the period 1/6/1999 to 29/6/2018. The S&P 500 Index represented the financial markets, EUR/USD and YEN/USD rep- resented the Forex markets, and eight types of commodities: Crude Oil, Natural Gas, Gasoline, and Propane represented energy commodities and Corn, Coffee, Wheat, and Soybeans represented food commodities. This analysis contribute to understanding of the dynamic frequency connectedness in case of a differentiated system of markets. The main finding was the strongest short-frequency reaction to shocks in case of all variables, which is opposite behavior than usually observed in banking sector frequency dynamics analyses. JEL Classication: F12, F21, F23, H25, H71, H87 Keywords: connectedness, financial market, forex market, commodity market, systemic risk, spillovers, frequency analysis Author's e-mail: Supervisor's e-mail:

National Repository of Grey Literature : 105 records found   1 - 10nextend  jump to record:
See also: similar author names
2 Baruník, Jozef,
Interested in being notified about new results for this query?
Subscribe to the RSS feed.