|
|
Neighborhood components analysis and machine learning
Hanousek, Jan ; Antoch, Jaromír (advisor) ; Maciak, Matúš (referee)
In this thesis we focus on the NCA algorithm, which is a modification of k-nearest neighbors algorithm. Following a brief introduction into classification algorithms we overview KNN algorithm, its strengths and flaws and what lead to the creation of the NCA. Then we discuss two of the most widely used mod- ifications of NCA called Fast NCA and Kernel (fast) NCA, which implements the so-called kernel trick. Integral part of this thesis is also a proposed algo- rithm based on KNN (/NCA) and Linear discriminant analysis titled TSKNN (/TSNCA), respectively. We conclude this thesis with a detailed study of two real life financial problems and compare all the algorithms introduced in this thesis based on the performance in these tasks. 1
|
|
|
Firm efficiency, foreign ownership and CEO gender in corrupt environments
Hanousek, Jan ; Shamshur, Anastasiya ; Trešl, Jiří
We study the effects of corruption on firm efficiency using a unique dataset of private firms from 14 Central and Eastern European countries from 2000 to 2013. We find that an environment characterized by a high level of corruption has an adverse effect on firm efficiency. This effect is stronger for firms with a lower propensity to behave corruptly, such as foreign-controlled firms and firms managed by female CEOs, while local firms and firms with male CEOs are not disadvantaged. We also find that an environment characterized by considerable heterogeneity in the perception of corruption is associated with an increase in firm efficiency. This effect is particularly strong for foreign-controlled firms from low corruption countries, while no effect is observed for firms managed by a female CEO.
|
|
|
Company names and their influence on company productivity
Hanousek, Jan ; Jurajda, Štěpán
Our analysis reveals that a company’s name may be related to its economic productivity. The name’s alphabetical position, the use of “national”4 words, and the inclusion of English phrases within the name all play a role. This study looks at the association between certain characteristics of a company’s name and growth in the company’s turnover, growth in value added and profit from shares using company data from 17 EU countries over a period of more than 15 years. We distinguish between companies in countries where Romance languages, Slavic languages and Germanic languages are spoken. The effects of company names vary across these language-country groups. The link between the alphabetical position of the name and the company’s economic productivity is strongest in the services sector. For example in France and Poland moving the company’s name from the end of the alphabet to the beginning can result in increased turnover by as much as two percentage points. The influence of “national” words in the company’s name is connected with higher long-term growth in turnover, in particular in Poland, France and Norway. In Poland the use of “national” words can increase long-term growth in turnover by nearly four percentage points. The influence of including English words in the company’s name varies across the languagecountry groups. While in the Romance group English words have a negative infuence, in Germanic group of countries they have mixed effects, and in Slavic language speaking countries they have a clear positive effect. This study adds to existing analyses in the field of marketing and linguistics that have focused in particular on the origin of brand and company names, and to research into the effects of company names on share profits.
|
|
|
Applications of modern spectral tools in financial econometrics
Křehlík, Tomáš ; Baruník, Jozef (advisor) ; Hanousek, Jan (referee) ; Croux, Christopher (referee) ; Wang, Yao (referee)
Spectral tools in econometrics have lately experienced a renewed surge in interest. This dissertation contributes to this literature by providing conceptually different spectral-based methods and their applications to problems of modern economics. In the first part, we take a spectral decomposition of realized volatility and construct a multivariate GARCH style model that we fit by standard quasi-maximum likelihood and generalized autoregressive score procedures. We build our model on a belief that market agents obtain information in various time horizons and therefore form their expectations in various informational horizons. This behavior creates an overall volatility process that is a mixture of spectrum specific processes. We then apply the model to the currency markets, namely GBP, CHF, and EUR. With the help of the model confidence set test we show that the multi-scale model and the generalized autoregressive score based models produce forecasts that are in most cases superior to the competing models. Moreover, we find that most of the information for future volatility comes from the high frequency part of the spectra representing the very short investment horizons. In the second part, we provide a spectral decomposition of a system multivariate connectedness measure based on Diebold and Yilmaz...
|
|
|
Productivity and trade spillovers: horizontal crowding-out versus vertical synergies in Europe as a response to foreign direct investment
Hanousek, Jan ; Kočenda, E. ; Vozárová, P.
We analyze the impact of multinational enterprises (MNEs), via their foreign direct investment (FDI), on domestic firms in 30 European host economies, from 2001 to 2013. We incorporate international industrial and trade linkages into a standard theoretical framework and test them empirically on a unique dataset compiled from the Amadeus, Eurostat, UN Comtrade and BACI data sources. While controlling for horizontal, vertical, and export channels at the upstream and downstream levels, we show that the presence of MNEs significantly affects domestic firms, in terms of both changing the market structure and improving productivity. The impact is not always positive, as domestic firms are often crowded-out. However, those firms that withstand such double competition receive additional benefits stemming from trade (export) spillovers. In our complex model, we did not find significant (positive) interactions of domestic firms with horizontal MNEs which would suggest desirable productivity spillovers.
|
|
|
Asymmetries in the firm’s use of debt to changing market values
Ferris, S. P. ; Hanousek, Jan ; Shamshur, Anastasiya ; Trešl, Jiří
Using a large sample of U.S. firms over the period, 1984 to 2013, this study examines the relation between market and book leverage ratios. Unlike Welch (2004) who contends that changes in market leverage do not induce adjustments in book leverage, we find an asymmetric effect. That is, firms adjust their book leverage relative to market leverage only when the changes in market leverage are due to increases in the value of the firm’s equity. No adjustment is observed when firm equity values decrease. We observe a number of interesting differences between those firms that make large and small capital structure adjustments in response to changing equity prices. Our results are consistent with Barclay, Morellec and Smith (2006) who argue that the optimal level of debt decreases in the presence of corporate growth options.
|
|
|
To bribe or not to bribe? Corruption uncertainty and corporate practices
Hanousek, Jan ; Shamshur, Anastasiya ; Trešl, Jiří
Using a large sample of private firms over the period from 2001 to 2013, we study the effect of corruption uncertainty on corporate investments and cash holdings. We find that a higher uncertainty about the level of corruption is associated with lower corporate investments and lower cash holdings. These results are sensitive to the ownership structure of a firm. Firms with no foreign majority ownership appear to be more sensitive to corruption-induced uncertainty than majority-controlled foreign firms. They significantly decrease their investments and cash holdings. We hypothesize that they move their cash off-balance-sheet to create cash reserves as the uncertainty of when, whom, and how much to bribe increases.
|
|
|
Essays on Public Finance
Lichard, Tomáš ; Hanousek, Jan (advisor) ; Palda, Filip (referee) ; Fialová, Kamila (referee)
This dissertation analyzes two presently widely discussed topics in Public Finance: relationship between the shadow economy and tax policy, and the effect of financial transaction taxes on the functioning of financial markets. The first chapter describes presently used estimators of the size of the shadow economy, with a focus on microeconomic estimators. It illustrates problems with assumptions that a vast majority of recent studies use to identify underreporting (mainly the comparison of employed and self-employed) using data from four transition economies as an example. It shows that the most common assumption, that self-employed evade whereas employees do not is probably too strict in less compliant economies, where even employees have opportunities to evade through e.g. under-the-table wages or by moonlighting at unreported jobs. The second chapter develops an estimator of unreported income that relaxes some of these strict assumptions. Assuming only that tax-evading households have a higher consumption- income gap than non-evaders in surveys, an endogenous switching model with unknown sample separation enables the estimation of both the probability of hiding income and the expected amount of unreported income for each household. Using data from Czech and Slovak household...
|
|
|
Essays on Financial Markets
Brushko, Iuliia ; Hanousek, Jan (advisor) ; Bredthauer, Jeffery (referee) ; Payne, Brian (referee)
English This thesis studies financial markets and the information we can obtain from observing the actions of financial market participants. In the first chapter, I study how the combination of different accounting ratios, which are considered to be the financial signals of future performance, can affect the analysts' and managers' earnings forecast releases. The findings show that analysts treat the firms differently depending on whether the firms have only strong financial indicators (high signal group), weak financial indicators (low signal group), and those with both positive and negative signals (mixed signal group). The study also provides the evidence that the managers may realize the heterogeneity in analysts' treatment, and as the result the managers' earnings forecasts will be affected both by the signal group type of the firm and the analysts' bias characteristic for the appropriate signal group. At the same time, the findings show that the analysts sometimes fail to disregard the managers' forecast biases and are misled by the managers. This provides evidence of inaccuracy on the part of analysts and potential gaming on information disclosure between analysts and managers'. In the second chapter, I examine whether trading activity responds to the industry-related earnings announcement...
|
|
|
Essays on Credit Risk
Kuvíková, Gabriela ; Hanousek, Jan (advisor) ; Fidrmuc, Jarko (referee) ; Podpiera, Jiří (referee)
The financial crises in the early 2000s have given prominence to the financial markets' exposure to credit risk. To minimize credit risk, the risk that a borrower will fail to meet her contractual obligations, lenders seek to identify borrowers with a high probability of default prior to granting credit. In my dissertation I examine several screening devices that lenders utilize in alleviating adverse selection present on the credit market. In the first chapter, I ask whether the existence of informal collateral signals better loan repayment. Taking advantage of a unique dataset of household loans from a Czech commercial bank, I find that housing loans without lien on the property default less compared to loans with unspecified purpose. I also show that the interest rate differential between specific purpose loans and unspecified purpose loans is systematically higher than their default rate differential. In the second chapter, I investigate the role of loan contract terms in household loan demand and performance. Utilizing a sample of accepted and rejected Czech household loans, I find that loan demand for low-income borrowers is more sensitive to liquidity constraint and loan maturity changes than to interest rate changes. The results also suggest that by reflecting the borrower's riskiness in...
|
guest ::
RSS feed.