|
|
Forecasting with neural network during covid-19 crisis
Luu Danh, Tiep ; Baruník, Jozef (advisor) ; Kukačka, Jiří (referee)
The thesis concerns the topic of forecasting using Neural Networks, particu- larly the return and volatility forecasting in the volatile period of Covid-19. The thesis uses adjusted close daily data from Jan 1, 2000, until Jan 1, 2021, of the S&P index and Prague Exchange Stock index (PX). The comparison was between the vanilla econometrical model, a neural network model, and a hybrid neural network model. Hybrid neural networks were constructed with an additional feature column of the fitted econometrical model. Additionally to comparing the prediction, a risk-return trade-o analysis of the forecasted series was conducted. The test period for all models was from Jan 1, 2020, until Jan 1, 2021, where predictions were made. During the test period, MSE be- tween predicted and true values was extracted and compared. The results are that the hybrid model outperformed both econometrical as well as only neural networks models. Furthermore, the risk-return trade-o forecast provided by the hybrid model fares better than the other ones. JEL Classification C53, C81 Keywords Financial Time Series, Forecasting, Neural Net- works, ARIMA, GARCH Title Forecasting with Neural Network during Covid- 19 Crisis Author's e-mail tiep.luud@gmail.com Supervisor's e-mail barunik@fsv.cuni.cz
|
|
| |
|
| |
|
|
Careless society: Drivers of (un)secure passwords
Nedvěd, Vojtěch ; Baruník, Jozef (advisor) ; Kukačka, Jiří (referee)
Careless Society: Drivers of (Un)Secure Passwords Thesis abstract Vojtěch Nedvěd May 2, 2021 Vulnerabilities related to poor cybersecurity are a dangerous global economic issue. This thesis aims to explain two examples of poor password management. First, why users use similar password and username and second, why they reuse their passwords, as the main drivers of this behaviour are unknown. We examined the effects of selected macroeconomic variables, gender, password length and password complexity. Additionally, this thesis suggest how to estimate sentiment in passwords using models build on Twitter posts. The results are verified on large password data, including password leaks from recent years. There are four main findings. First, a higher cybersecurity index and diversity of a password seem to be related to the lower similarity between a username and a password. Second, it seems that there are structural differences between countries and languages. Third, the sentiment seems to be a significant determinant too. Fourth, password reuse seems to be positively affected by the cybersecurity level. The thesis contributes to the study of password management. It proposes how to model the relationship, derive the data, split the passwords into words, model the sentiment of passwords, what variables might be...
|
|
|
Evaluating the predictability of virtual exchange rates using daily data
Řanda, Martin ; Polák, Petr (advisor) ; Kukačka, Jiří (referee)
Virtual worlds have garnered the attention of researchers from various disci- plines and are viewed as particularly valuable to economists due to their open- ended design. In this thesis, we review a popular online multiplayer game's economy and focus on exchange rate predictability in a virtual setting as only a limited body of literature investigated this topic. The well-established unpre- dictability puzzle is addressed by exploiting a unique daily time series dataset using a vector autoregressive framework. Apart from a significant Granger- causal relationship between the virtual exchange rate and the player popula- tion, the system is shown to be less interconnected than expected. Furthermore, an out-of-sample exercise is conducted, and the forecasting performance of our models is examined in comparison to that of a simple no-change benchmark in the short term. Based on the evaluation methods used, the two measures of the virtual exchange rate are found to be somewhat predictable. We suggest two explanations for this inconsistency between the virtual and real-world exchange rates: data frequency and lack of complexity in the considered online economy.
|
|
|
Is hype really that powerful? The correlation between mass and social media and cryptocurrency rates fluctuations
Ilina, Viktoriia ; Král, Michal (advisor) ; Kukačka, Jiří (referee)
Twelve years after Satoshi Nakamoto published the paper describing the functioning mechanism and principals of cryptocurrency that maintains secure and anonymous digital transactions beyond any banks, cryptocurrencies have become a multi-billion-dollar industry comprising millions of investors, miners, developers and profiteers. However, the actual price determinants and ways to forecast future price changes remain an open question yet to discover the answer for. This study attempts to figure out whether media hype exerts that much influence upon cryptocurrencies price movements and whether it can be used as the basis for future movements prediction. Two cryptocurrencies, Bitcoin and Tezos, and 7 mass and social media factors for each of them were considered on daily basis from 08-01-2018 to 10-31-2020. To explore the interdependence between media drivers and cryptocurrencies' prices in short, medium and long timespan, this study deploys wavelet coherence approach. There was found, that price changes turn to be the supreme prior to hype, even though the growing ado may push the prices even higher. Thus, hype is failing to prove itself as a reliable cryptocurrency price predictor. Crypto investors, though, should anyways take the news background into account while building trading strategies,...
|
|
|
How much do we pay for a real estate ownership? A simulation approach
Gallová, Ivana ; Pleticha, Petr (advisor) ; Kukačka, Jiří (referee)
This thesis compares rent vs home ownership based on the net present value, within the periods of the Great Recession and current year. The analysis is focused on the Czech Republic real estate market as a whole. Rent and real estate price are forecasted, and factors determining the price of rent and real estate are identified. The ARIMA model used for forecasting performs accurate short-term predictions. The results expect 3,2 percent annual growth of rent in the following year and 7,2 percent increase for the real estate prices. The results of net present value analysis indicate, that for years 2008 and 2009 renting was superior choice, while for years 2011, 2013 and 2019 home ownership was to be preferred from financial aspect.
|
|
|
Asset Prices, Network Connectedness, and Risk Premium
Procházková, Vendula ; Baruník, Jozef (advisor) ; Kukačka, Jiří (referee)
This diploma thesis introduces the measures of network connectedness in the context of asset pricing. It proposes an asset pricing model in which the factor of connectedness is included as one of the risk factors together with the three Fama-French factors. The goal of the analysis is to examine whether the con- nectedness represents a signifcant risk factor that should be considered while determining the risk premium of the portfolio in diferent sectors in the market. Using the realized volatilities and returns of 496 assets of SP 500 index over the period 2005 - 2018, that are divided into 11 sectors, we frstly determine the linkages of connectedness between the assets in the same sector. Applying Fama-MacBeth two-step regression model, we explore the signifcance of the connectedness factor for the determination of the risk premium. We argue that the sector overall connectedness represents a signifcant risk in most of the sec- tors and should be therefore taken into account by the investors in all sectors. Moreover, the total directional connectedness that captures the spillover of shocks to one asset from the other assets in the sector, is a signifcant risk fac- tor that should increase the risk premium of the portfolio, especially in sectors such as the fnancial, health care, consumer...
|
|
|
The Effect of M&A on Competitors' Performance in China and the US
Wojnarová, Renáta ; Kukačka, Jiří (advisor) ; Teplý, Petr (referee)
We examine the effect of merger announcements on the stock performance of acquirers' industry rivals in the context of Chinese and US deals between 1994 and 2017. Our analysis reveals that investors of rivals are able to earn abnormal returns during days around merger announcement, meaning that markets are not fully efficient as implied by the Efficient market hypothesis. We conclude that in a reaction to the announcement, US rivals achieve generally negative abnormal returns with higher magnitude and volatility compared to Chinese rivals. Additionally, we observe that Chinese investors' perception of mergers turned out to be more conservative after the Global financial crisis. During days around the merger announcement, signs of rivals' abnormal returns also differ on whether the target is public or private in both countries. Rivals operating in industries that are substantially supported by Chinese government such as real estate, pharmaceuticals, and chemicals experience positive reaction on mergers of their competitors. Furthermore, we find that industries with increasing im- portance in Chinese developing economy such as banking, telecommunications, and cyclical consumer products show a positive reaction of rivals' returns on merger announcements while in the developed US economy, a negative...
|
|
|
Predicting stock market crises using investor sentiment indicators
Havelková, Kateřina ; Kukačka, Jiří (advisor) ; Kočenda, Evžen (referee)
Using an early warning system (EWS) methodology, this thesis analyses the predictability of stock market crises from the perspective of behavioural fnance. Specifcally, in our EWS based on the multinomial logit model, we consider in- vestor sentiment as one of the potential crisis indicators. Identifcation of the relevant crisis indicators is based on Bayesian model averaging. The empir- ical results reveal that price-earnings ratio, short-term interest rate, current account, credit growth, as well as investor sentiment proxies are the most rele- vant indicators for anticipating stock market crises within a one-year horizon. Our thesis hence provides evidence that investor sentiment proxies should be a part of the routinely considered variables in the EWS literature. In general, the predictive power of our EWS model as evaluated by both in-sample and out-of-sample performance is promising. JEL Classifcation G01, G02, G17, G41 Keywords Stock market crises, Early warning system, In- vestor sentiment, Crisis prediction, Bayesian model averaging Title Predicting stock market crises using investor sentiment indicators
|
guest ::
