|
|
Conditional quantile models for asset returns
Havel, Štěpán ; Baruník, Jozef (advisor) ; Fanta, Nicolas (referee)
The literature related to Value at Risk estimation is rich in general. However, majority of papers written on this subject concentrates on the unconditional non-parametric or parametric approach to VaR modelling. This thesis focuses on direct conditional VaR estimation using quantile regression. Thereby im- posing no restrictions on the return distribution. We use daily volatility mea- surements for individual stocks in S&P 500 index and quantile regress them on one-day ahead returns of the entire index. Depending on the quantile selected this estimation produces different confidence levels of Value at Risk. In order to minimize complexity of the final model, regularization methods are applied. To the author's knowledge this specific methodology has not yet been applied in any paper. The main objective is to investigate whether this approach is able to produce sound VaR estimates comparable with different methods usu- ally applied. Our result suggests that quantile regression extended with lasso regularization can be used to produce sound one-day-ahead Value at Risk es- timates. JEL Classification C22, C58, G15 Keywords volatility, quantile regression, VaR, GARCH Title Conditional quantile models for asset re- turns Author's e-mail havel.stepan@gmail.com Supervisor's e-mail barunik@fsv.cuni.cz
|
|
|
Non-Performing Loans - Determinants, the Development over Time and the Impact on Banks and the Real Economy
Kafková, Kateřina ; Pečená, Magda (advisor) ; Fanta, Nicolas (referee)
Non-Performing Loans - Determinants, Development over Time and the Impact on Banks and the Real Economy Author: Kateřina Kafková Abstract This thesis explains the concept of non-performing loans (NPL) and analyses factors determining the share of NPLs in total gross loans provided in the Czech Republic. A panel of 24 banks operating in the Czech Republic with annual data from 2010-2019 is analysed. The main estimation method that is used is the difference Generalized Method of Moments. The possible determinants that are examined come from both macroeconomic and banking environment. The results of the estimation provide evidence of the existence of a connection between the NPL ratio and the macroeconomic factors, of which the effect of inflation and unemployment was the most significant. Also, the estimation confirms that the NPL ratio is significantly influenced by the bank-specific determinants, specifically by the effect of the previous values of the NPL ratio and the effect of credit growth. Finally, the thesis discusses the reversed effect - the effect of NPLs on the real economy.
|
|
|
Sustainable Development Goal Nr. 1: End of Poverty
Komorová, Anežka ; Cahlík, Tomáš (advisor) ; Fanta, Nicolas (referee)
Nowadays, the question of the Sustainable Development Goals (SDGs) seems to get more and more attention. This bachelor thesis focuses on the first goal - end poverty in all its forms everywhere. There is a great number of literature which examines possible determinants of poverty in order to create effective strategies to fight witht it. Majority of them focuses on the study in particular countries or regions since their data are easier to access. On the contrary, this thesis examines variables and their effect on extreme poverty in the world with the use of the econometric analysis on panel data. There are seven independent variables used: GDP per capita, Gini index as a proxy variable for income inequality, population growth and four education attainment levels - no education, primary, secondary and tertiary. The World Bank was chosen as the primary source of data with a great number of observations from 125 countries over time period of 2000 and 2017. According to the results of the study, explanatory variables GDP per capita, income inequality, no education and tertiary education significantly affect extreme poverty. Also, the results show that the goal of 0 % is unlikely to be fulfilled by the year of 2030. To sum up, the first goal of the SDGs turned out as too ambitious. For its...
|
|
|
The impact of macroeconomic news announcements on the value and volatility of selected foreign exchange rates in EU
Bubniak, Peter ; Fanta, Nicolas (advisor) ; Krištoufek, Ladislav (referee)
Bibliographic note BUBNIAK, Peter. The impact of macroeconomic news announcements on the value and volatility of selected foreign exchange rates in EU. Prague 2019. 47 pp. Bachelor thesis (Bc) Charles University, Faculty of Social Sciences, Institute of Economic Studies. Thesis supervisor: Mgr Nicolas Fanta. Abstract This work analyzes the influence of positive and negative macroeconomic news on the value of exchange rate and volatility. We have chosen EUR/USD, EUR/CZK and USD/CZK as our exchange rates. The influence of macroeconomic news published by Czech national bank and European central bank were analysed. For our purposes were used econometric models GARCH(1,1) and EGARCH(1,1) with both Normal and Student's distribution of error terms. One of the major outcomes were the importance of macroeconomic news on value and volatility on the exhcange rates. For each exchange rate has effect different macroeconomic index. The crucial are: Consumer price index and Harmonised Index of Consumer Pirces, unemplyoment rate and PRIBOR and EURIBOR. Another conclusion was that our financial dataset displays the main nature of volatility. JEL Classification C22, E00, E52, E58, F3, F4, F31, G1, G13, G14 Key words financial market, exchange rate, ARCH model, GARCH model, volatility Authors e-mail bubniak.peter@gmail.com...
|
|
|
Neural networks and tree-based credit scoring models
Turlík, Tomáš ; Krištoufek, Ladislav (advisor) ; Fanta, Nicolas (referee)
The most basic task in credit scoring is to classify potential borrowers as "good" or "bad" based on the probability that they would default in the case they would be accepted. In this thesis we compare widely used lo- gistic regression, neural networks and tree-based ensemble models. During the construction of neural network models we utilize recent techniques and advances in the field of deep learning, while for the tree-based models we use popular bagging, boosting and random forests ensembling algorithms. Performance of the models is measured by ROC AUC metric, which should provide better information value than average accuracy alone. Our results suggest small or even no difference between models, when in the best case scenario neural networks, boosted ensembles and stacked ensembles result in only approximately 1%−2% larger ROC AUC value than logistic regression. Keywords credit scoring, neural networks, decision tree, bagging, boosting, random forest, ensemble, ROC curve
|
|
|
ECB's Oral Communication and Future Monetary Policy
Fanta, Nicolas ; Horváth, Roman (advisor) ; Schneider, Ondřej (referee)
i A bstr a ct T he t hesis ai ms t o s he d li g ht o n t he E ur o p e a n Ce ntr al B a n k s c o m m u nic ati o n i n or der t o i de ntif y its m ai n c o m p o ne nts b e ari n g i nf or m ati o n a b o ut f ut ure c h a n ges i n t he p olic y r ate. For t he a n al ysis, t he st u d y i ntr o d uces a m o di fic ati o n of a wi del y use d a p pr o ac h b ase d o n t he disse nt e x presse d i n pre vi o us m o net ar y p olic y v ote. Si nce t he E ur o p e a n Ce ntr al B a n k d o es n ot p u blis h t he v ote's det ails t he c o m m u nic ati o n of t he ce ntr al b a n k is use d as a pr o x y. Res ults n ot o nl y c o n fir m t he pre dicti ve p o wer of t he c o m m u nic ati o n b ut f urt her m ore i n dic ate t h at t he fi n a nci al m ar kets d o n ot f ull y i nc or p or ate t he i nf or m ati o n c o nt ai ne d. A det aile d a n al ysis s h o ws t he rele va nce of t he ti mi n g, deli ver y a n d c o nte nt of t he c o m m u nic ati o n. T he st u d y t heref ore pr o vi des a s u m m ar y of t he i m p ort a nt f act ors of t he E ur o p e a n Ce ntr al B a n k b o ar d me m b ers st ate me nts f or pre dicti n g f ut ure m o net ar y p olic y.
|
|
|
Price Elasticity of Alcohol Demand: A Meta-Analysis
Fanta, Nicolas ; Havránek, Tomáš (advisor) ; Zeynalov, Ayaz (referee)
The own-price elasticity is considered to be one of the key factors describing the demand for alcohol. There have been many estimates computed by now but only a few studies tried to analyse them. The aim of this meta-analysis is to discover more about the eventual effects that publication bias might have in the alcohol-related literature. The first part describes the various types of elasticities and the methods of estimation. This study is estimating the so called true effect elasticity in order to show how elastic the demand for alcoholic beverages is. As there are many ways how to estimate the elasticities it is also analysed if different approaches to the estimation lead to different results. The use of modern meta- analytical methods leads to significantly different results from the ones of previous meta-analyses. The estimated true effects yields new evidence that the demand for alcoholic beverages might be perfectly inelastic. Evidence of publication bias is quite strong and it appears that the economics research cycle hypothesis is also valid.
|
guest ::
