Buuruljin Enkhbold Finance and Growth Nexus: CEE & Central Asia and Beyond Abstract (English) This thesis investigates the effect of financial development on economic growth using both global sample and regional samples focusing on Central and Eastern Europe (CEE) and Central Asia during the time period 1960-2013. The results of fixed effect panel and system GMM estimators suggest that the effect of private credit on growth had been neutral until 2007 and the effect turns negative if the time period is up to 2013. The negative effect of private credit on growth has been the largest for CEE and Central Asia, particularly for non-EU countries in the region. Stock market capitalisation and lending deposit spread have consistent effects regardless of the choice of time frame which implies that economies benefit from larger stock markets and lower lending deposit spread. Keywords: financial development, credit, stock market, spread, growth, CEE and Central Asia, generalized method of moments (GMM)

Real estate market prices have proven to be influenced by many driving factors. As suggested by Tiebout's (1956) model, the level of public services was believed to influence people's decision to move geographically, which was called "voting with their feet". The panel data used in this paper were the market price of one- or two- dwelling buildings and the per capita cost of public investment factors in 290 municipalities in Sweden for the period 2011 to 2020. The principal component analysis method was firstly performed to dimensionalize the data and create new index variables. A panel vector autoregressive model based on the GMM estimator was developed, and cluster robust standard errors were used to address autocorrelation and heteroskedasticity. The empirical results showed that in addition to demographic variables, the principal components of government investment variables in traffic infrastructure, youth education and comprehensive health care, had a positive stimulating effect on real estate prices. The results of Granger causality tests showed a positive effect of the increment in house prices on investment in public activities. The IRF plots and variance decomposition of the first-order difference in real estate prices over the next ten periods showed that despite the presence of positive stimulus for public investment, house prices were more independent in the future, which indicated the dominant factor in the movements was itself. / Priserna på fastighetsmarknaden har visat sig påverkas av många faktorer. Enligt Tiebouts (1956) modell trodde man att nivån på de offentliga tjänsterna påverkade människors beslut att flytta geografiskt, vilket kallades "rösta med fötterna". De paneldata som används i denna artikel är marknadspriset på en- eller tvåbostadshus och kostnaden per capita för offentliga investeringsfaktorer i 290 kommuner i Sverige för perioden 2011-2020. Först utfördes PCA-metoden för att dimensionera data och skapa nya indexvariabler. En auktoregressiv panelvektormodell baserad på GMM-estimatorn utvecklades, och klusterrobusta standardfel användes för att hantera autokorrelation och heteroskedasticitet. De empiriska resultaten visade att förutom demografiska variabler hade huvudkomponenterna i de statliga investeringsvariablerna i trafikinfrastruktur, ungdomsutbildning och omfattande hälsovård en positiv stimulerande effekt på fastighetspriserna. Resultaten av Granger-kausalitetstesterna visade att ökningen av huspriserna har en positiv effekt på investeringar i offentlig verksamhet. IRF-plottarna och variansdekompositionen av första ordningens skillnad i fastighetspriserna under de följande tio perioderna visade att trots positiva stimulansåtgärder för offentliga investeringar var huspriserna mer oberoende i framtiden, vilket tydde på att den dominerande faktorn i rörelserna var sig själv.

This thesis investigates the computer vision application of live fish recognition, which is needed in application scenarios where manual annotation is too expensive, when there are too many underwater videos. This system can assist ecological surveillance research, e.g. computing fish population statistics in the open sea. Some pre-processing procedures are employed to improve the recognition accuracy, and then 69 types of features are extracted. These features are a combination of colour, shape and texture properties in different parts of the fish such as tail/head/top/bottom, as well as the whole fish. Then, we present a novel Balance-Guaranteed Optimized Tree with Reject option (BGOTR) for live fish recognition. It improves the normal hierarchical method by arranging more accurate classifications at a higher level and keeping the hierarchical tree balanced. BGOTR is automatically constructed based on inter-class similarities. We apply a Gaussian Mixture Model (GMM) and Bayes rule as a reject option after the hierarchical classification to evaluate the posterior probability of being a certain species to filter less confident decisions. This novel classification-rejection method cleans up decisions and rejects unknown classes. After constructing the tree architecture, a novel trajectory voting method is used to eliminate accumulated errors during hierarchical classification and, therefore, achieves better performance. The proposed BGOTR-based hierarchical classification method is applied to recognize the 15 major species of 24150 manually labelled fish images and to detect new species in an unrestricted natural environment recorded by underwater cameras in south Taiwan sea. It achieves significant improvements compared to the state-of-the-art techniques. Furthermore, the sequence of feature selection and constructing a multi-class SVM is investigated. We propose that an Individual Feature Selection (IFS) procedure can be directly exploited to the binary One-versus-One SVMs before assembling the full multiclass SVM. The IFS method selects different subsets of features for each Oneversus- One SVM inside the multiclass classifier so that each vote is optimized to discriminate the two specific classes. The proposed IFS method is tested on four different datasets comparing the performance and time cost. Experimental results demonstrate significant improvements compared to the normal Multiclass Feature Selection (MFS) method on all datasets.

本文採用1980-2005年共94國的縱橫資料(Panel data)，探討金融發展與經濟成長的關係。有別於一般傳統採用單一迴歸模型的架構，本文改採聯立迴歸模型探討兩者之因果關係。 本研究分別採用四個「金融發展指標」及三個「股市發展指標」，輔以與金融發展和經濟成長密切相關的外生控制變數，再分別與經濟成長指標建構聯立迴歸模型。此外，有關參數之估計與認定，以Lewbel (2007)的計量方法為立論依據，在殘差項具有異質共變異現象的限制條件下，利用一般化動差法(GMM)，同時估計聯立方程式的結構式參數。是故，本文根據不同之金融發展與股市發展變數的設定，總共建立七個聯立方程式模型，分別探討金融發展、股市發展與經濟成長之交互影響效果。實證結果彙整成以下兩點說明： 一、 有關金融發展指標對經濟成長之綜合影響的實證結果顯示，惟有「存款貨幣機構在本國資產總額/(存款貨幣機構在本國資產總額＋央行本國資產總額)」對經濟成長具有正向的綜合影響，其他三項金融發展指標對經濟成長皆無顯著影響；另一方面，除了以「銀行部門對其本國之總放款額/GDP」做為金融發展指標的模型4為不顯著外，模型1至模型3皆顯示經濟成長分別對「存款貨幣機構在本國資產總額/(存款貨幣機構在本國資產總額＋央行本國資產總額)」、「全體金融機構流動負債/GDP」以及「全體金融機構對私部門放款總額/GDP」等三項金融發展指標存在正向綜合影響。 二、 有關本研究股市發展指標對經濟成長之綜合影響效果的實證結果分歧。其中，「股市總成交金額/股市總市值 (股市週轉率)」對經濟成長有正向綜合影響；「上市公司股票總市值/GDP (股市資本率)」對經濟成長有負向綜合影響；而「股市總成交金額/GDP (股市總成交值比)」對經濟成長則無顯著之綜合影響；另一方面，考量經濟成長之當期項與落後期的綜合效果對股市發展的影響，本研究發現「經濟成長變數」的當期項與落後期對三個股市發展指標均無顯著的綜合影響。

This thesis addresses the question, whether the impact of financial development on the relative and absolute indicators of poverty is dependent on the levels of the human capital present in an economy. To answer this question, first we develop a theoretical framework to explain the growth process in the context of financial development assuming that human capital is heterogeneous in terms of the skills and education people have. Then, by using the data sets based on five-year averages over 1960-2010 and 1980-2010, covering 107 developed and developing countries, we empirically investigate the extensions of the theoretical framework developed earlier. These extensions cover the relationships between: 1. Income inequality and economic growth 2. Financial development, human capital and income inequality, and 3. Financial development, human capital and poverty We provide empirical evidence using modern panel data techniques of dynamic and static GMM. The findings elucidate that income inequality and economic growth are inter-dependent on each other. There exists an inverse relationship between initial inequality and economic growth. The changes in income inequality follow the pattern identified by Kuznets (1955) known as Kuznets’ hypothesis. The results also show that financial development helps in reducing income inequalities and in alleviating poverty, only when there is a sufficient level of human capital available. On the basis of our findings we develop the term "effective financial development" which means that financial development is effective in accelerating growth levels, reducing income inequalities and alleviating poverty only if there is a sufficient level of human capital available. The empirical study covers multiple aspects of financial development like private credit extended by banks and other financial institutions, liquid liabilities and stock market capitalization. The results of the empirical investigations are robust to multiple data sets and various indicators of income inequality, financial development, poverty and human capital. The study also provides marginal analysis, which helps in understanding the impact of financial development on inequality and poverty at different levels of human capital. This research study of effective financial development can be a useful learning paradigm for the academics and researchers interested in growth economics and keen to learn how poverty and income inequality can be reduced effectively. This study can also be useful for the policy makers in the financial institutions, because it provides robust empirical evidence that shows that financial development cannot help in alleviating poverty and in reducing inequalities unless there is a sufficient level of human capital available. The findings can be useful for policy makers, particularly in the developing countries where high levels of income inequalities and poverty are big problems. This study explains the mechanism of how effective financial development can be used to reduce income inequalities and to alleviate poverty. It also explains the process of inter-linkages between financial development, human capital, inequality, economic growth and financial instability. The policy makers can also take advantage from the marginal analyses that illustrate the minimum levels of private credit and primary and secondary schooling above which the effects of financial development and human capital become significant in reducing inequalities and poverty.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.

This thesis deals with both theoretical and practical aspects of banking fee and commission income in the European Union. Since fee income represents the largest part of non-interest income earned by banks, it remains a major challenge for bank management to set and maintain an appropriate fee policy. Nevertheless, solving for the optimal fee structure has not yet been accomplished either on a theoretical level, or in actual practice. In the thesis, we analyse fee income in EU banking sectors. Our results show that the Czech banking sector was not abnormally dependent on fee income compared to other EU countries in the period 2007-2012. As a result, we argue that the high profitability of Czech banks cannot be attributed to abnormal banking fees and commission income, but rather other factors should be considered. Moreover, we study the determinants of fee income share in individual banks and discuss the impact of market concentration on the magnitude of banking fees. We conclude that banks facing higher competition tend to expand more aggressively into non- traditional activities and therefore they report higher fee income shares. We also study the relationship between banking fees and banks' performance. The results are mixed depending on applied profitability measure, but in general, banks with...

This thesis is interested in relative performance of highly criticized Czech credit unions. Theoretical part comments on their historical development, makes international comparison and shows possible development of legislation. We created two unique datasets to assess financial performance of Czech credit unions in subsequent empirical part. The first one contains Czech credit unions' and commercial banks' data. The second one is established to make a comparison of Czech credit unions with cooperative banks operating elsewhere in the EU. Both are based on annual data between 2007 and 2012 period. System GMM method is employed as main instrument of our empirical analysis and alternative panel data methods are used as supplementary techniques. We focused our analysis on comparison of relative profitability and stability measures of Czech credit unions. The results revealed their poor performance in the given time period. According to our estimates, they resembled rather small commercial banks than cooperative ones. The negative relationship between Czech credit unions' stability measure (Z-score) and their asset size is especially striking. Moreover, Z- score of Czech credit unions decreased sharply in 2012. Such development was observed neither in case of Czech commercial banks nor in other...

This paper estimates the determinants of exchange rate variability for 21 developed economies in 1980-1998. The results show that traditional criteria implied by the optimum currency area (OCA) theory, such as business cycle synchronisation, trade linkages and economy size, determine to a large extent bilateral exchange rate variability. Using the ordinary least squares estimation, we compute OCA indices for European economies vis-à-vis Germany and identify countries showing consistently large or little signs of convergence. We find that since 1998, most European developed economies have converged to Germany whether or not they are using the euro, suggesting that structural similarity is not driven solely by monetary integration. Our results from the model estimated by the generalized method of moments suggest that two additional criteria reflecting labour market flexibility and private credit growth are significant in explaining the exchange rate variability and lead to a ranking of countries different from the traditional approach. We find a positive relationship between the OCA indices and GDP decline during the economic crisis of 2008-09, which further supports the view that the OCA index is a useful indicator of the candidates' readiness to join the Euro Area. We apply the results to the...

Acoustic analysis of animal vocalizations has been widely used to identify the presence of individual species, classify vocalizations, identify individuals, and determine gender. In this work automatic identification of speaker and gender of mice from ultrasonic vocalizations and speaker identification of meerkats from their Close calls is investigated. Feature extraction was implemented using Greenwood Function Cepstral Coefficients (GFCC), designed exclusively for extracting features from animal vocalizations. Mice ultrasonic vocalizations were analyzed using Gaussian Mixture Models (GMM) which yielded an accuracy of 78.3% for speaker identification and 93.2% for gender identification. Meerkat speaker identification with Close calls was implemented using Gaussian Mixture Models (GMM) and Hidden Markov Models (HMM), with an accuracy of 90.8% and 94.4% respectively. The results obtained shows these methods indicate the presence of gender and identity information in vocalizations and support the possibility of robust gender identification and individual identification using bioacoustic data sets.