The gender wage gap in the labour market has long been a topic of study, highlighting the disadvantages faced by women in terms of earningscompared to men. This study aims to investigate if the Great Recession had additional impacts on women's earnings differentials. Using census data from 2006 to 2012 in the US, two different quantile regressions were conducted for various income quantiles. One regression excluded variables, while the other included socio-demographic characteristics. The results indicate clear wage differences for women before, during, and after the Great Recession.The first regression shows statistically significant negative correlations between logarithmic income and gender. The quantile regressions also reveal decrease in the gender wage gap during the recession, with education returns favouring women in 2008 and 2009 before returning to pre-recession levels. Additionally, the results suggest that married women and women with children tend to have lower earnings compared to their male counterparts.These findings provide evidence of a glass ceiling in the US labour market, which may have been exacerbated by the exogenous shock of the Great Recession.

With a background in EU regulations and an increased interest in Environmental, Social, and Governence (ESG) policies in companies when investing, this thesis considers the individual contributions to market risk in portfolios by different ESG parameters. It explores two different methods to examine if there are effects consistent across the whole Nordic markets, and the possibility to express any effects within portfolios in a clear way. It uses the OMXNORDIC index as the market index and two different fund portfolios as example portfolios, one of which is an article 9 fund. The quantile regression approach does not show any consistent effects across the whole Nordic market from any ESG parameter explored. It does however make for a clear way to present the effects on the portfolio level for each ESG parameter. The employed Copula approach does show some consistent difference between the ESG parameters for the market and in portfolios, as well as differences between the portfolios. Both of the explored methods should allow for comparisons between, and reports on, fund portfolios which would improve the ESG analyses of funds. / Mot bakgrund av EU-lagstiftning och ett ökat intresse i företags förhållning till Environmental, Social, och Governence (ESG) frågor, så utforskar den här uppsatsen ESG-faktorers bidrag till marknadsrisk i fondportföljer och på den nordiska marknaden. Uppsatsen använder två olika metoder för att undersöka om det finns potentiella konsekventa effekter på den Nordiska aktiemarknaden, och möjligheten att presentera resultat på portföljnivå på ett tydligt sätt. OMXNORDIC används som marknadsindex, och två olika fondportföljer används som exempelportföljer, varav en är en artikel 9 fondportfölj. Quantile regression-metoden visar inte på några konsekventa effekter över hela den nordiska marknaden, för någon av ESG-parametrarna. Däremot så resulterar metoden i ett tydligt sätt att presentera påverkan av ESG-parametrarna på portföljnivå. Copula-metoden som används visar på några konsekventa skillnader mellan ESG-parametrar, både för marknaden och i fondportföljerna, samt skillnader mellan portföljerna i sig. Båda metoderna lämpar sig till att jämföra och bygga rapporter på fondportföljer, vilket borde leda till bättre ESG-analyser av fonder.

Time series anomaly detection receives increasing research interest given the growing number of data-rich application domains. Recent additions to anomaly detection methods in research literature include deep learning algorithms. The nature and performance of these algorithms in sequence analysis enable them to learn hierarchical discriminating features and time-series temporal nature. However, their performance is affected by the speed at which the time series arrives, the use of a fixed threshold, and the assumption of Gaussian distribution on the prediction error to identify anomalous values. An exact parametric distribution is often not directly relevant in many applications and it’s often difficult to select an appropriate threshold that will differentiate anomalies with noise. Thus, implementations need the Prediction Interval (PI) that quantifies the level of uncertainty associated with the Deep Neural Network (DNN) point forecasts, which helps in making a better-informed decision and mitigates against false anomaly alerts. To achieve this, a new anomaly detection method is proposed that computes the uncertainty in estimates using quantile regression and used the quantile interval to identify anomalies. Similarly, to handle the speed at which the data arrives, an online anomaly detection method is proposed where a model is trained incrementally to adapt to the concept drift that improves prediction. This is implemented using a window-based strategy, in which a time series is broken into sliding windows of sub-sequences as input to the model. To adapt to concept drift, the model is updated when changes occur in the new arrival instances. This is achieved by using anomaly likelihood which is computed using the Q-function to define the abnormal degree of the current data point based on the previous data points. Specifically, when concept drift occurs, the proposed method will mark the current data point as anomalous. However, when the abnormal behavior continues for a longer period of time, the abnormal degree of the current data point will be low compared to the previous data points using the likelihood. As such, the current data point is added to the previous data to retrain the model which will allow the model to learn the new characteristics of the data and hence adapt to the concept changes thereby redefining the abnormal behavior. The proposed method also incorporates feature extraction to capture structural patterns in the time series. This is especially significant for multivariate time-series data, for which there is a need to capture the complex temporal dependencies that may exist between the variables. In summary, this thesis contributes to the theory, design, and development of algorithms and models for the detection of anomalies in both static and evolving time series data. Several experiments were conducted, and the results obtained indicate the significance of this research on offline and online anomaly detection in both static and evolving time-series data. In chapter 3, the newly proposed method (Deep Quantile Regression Anomaly Detection Method) is evaluated and compared with six other prediction-based anomaly detection methods that assume a normal distribution of prediction or reconstruction error for the identification of anomalies. Results in the first part of the experiment indicate that DQR-AD obtained relatively better precision than all other methods which demonstrates the capability of the method in detecting a higher number of anomalous points with low false positive rates. Also, the results show that DQR-AD is approximately 2 – 3 times better than the DeepAnT which performs better than all the remaining methods on all domains in the NAB dataset. In the second part of the experiment, sMAP dataset is used with 4-dimensional features to demonstrate the method on multivariate time-series data. Experimental result shows DQR-AD have 10% better performance than AE on three datasets (SMAP1, SMAP3, and SMAP5) and equal performance on the remaining two datasets. In chapter 5, two levels of experiments were conducted basis of false-positive rate and concept drift adaptation. In the first level of the experiment, the result shows that online DQR-AD is 18% better than both DQR-AD and VAE-LSTM on five NAB datasets. Similarly, results in the second level of the experiment show that the online DQR-AD method has better performance than five counterpart methods with a relatively 10% margin on six out of the seven NAB datasets. This result demonstrates how concept drift adaptation strategies adopted in the proposed online DQR-AD improve the performance of anomaly detection in time series. / Petroleum Technology Development Fund (PTDF)

No description available.

本文以臺灣八大類股指數結合分量迴歸模型進行價量關係研究。有別於過去文獻多使用大盤指數進行分析，本文將以產業類股指數作為研究目標。實證結果顯示 : 「價量背離」與「價量齊揚」的效果同時存在於臺灣股市各個類股的價量關係中，且後者的效果普遍高於前者；而在八個產業類股中，尤以金融業在兩側分量的效果大於其他產業。另外，在相同的交易機制下，並非所有產業的價量關係皆會受到漲跌幅限制的影響而改變。本文更進一步選用法人持股佔該類股市值比作為資訊不對稱之代理變數，結果發現資訊不對稱程度較高的產業，在價量齊揚時，法人持股比的係數為負，代表在市場出現正報酬時，會有抑制股價上揚的效果；反之，在負報酬時，會加深股價下挫的力道。 / This research examines the relation between stock return and trading volume of Taiwan’s eight industries using quantile regression model. Our empirical results show that, for most industry indices, both large positive returns and large negative returns are usually accompanied by a large trading volume, with the effect of large positive returns being stronger. Among all industries, the financial industry has the most significant effect in either situation. But for some industries, the price-volume relations change when returns approach the price limits. In addition, we also emphasize the impact of information asymmetry, using ownership share of institutional investors as the proxy variable. The results show that, in the situation of positive returns with large trading volume, the institutional trading variable will restrain stock price from continually rising. In contrast, in the situation of negative returns with large trading volume, the institutional effect will make the stock price overreact.

No description available.

No description available.

The general focus of this thesis is on how the family, work and the welfare system are intertwined. A major determinant is the way responsibilities are shared by the state, the market and civil society in different welfare state regimes. An introductory chapter will therefore be dedicated to the development of the social dimension in the process of European integration. A first chapter will then go deeper into the comparative analysis of welfare state regimes, to comment on the provision of welfare in societies with a different mix of state, market and societal welfare roles and to assess the adequacy of existing typologies as reflections of today’s changed socio-economic, political and gender reality. Although they stand strong on their own, these first two chapters also contribute to contextualising the research subject of the remainder of the thesis: the study and comparison of the differential situation of women and men and of mothers and non-mothers on the labour markets of the EU-15 countries as well as of the role of public policies with respect to the employment penalties faced by women, particularly in the presence of young children. In our analysis, employment penalties are understood in three ways: (i) the difference in full-time equivalent employment rates between mothers and non-mothers, (ii) the wage penalty associated with motherhood, and (iii) the wage gap between part-time and full-time workers, considering men and women separately. Besides from a gender point of view, employment outcomes and public policies are thus assessed comparatively for mothers and non-mothers. Because women choose to take part in paid employment, fertility rates will depend on their possibilities to combine employment and motherhood. As a result, motherhood-induced employment penalties and the role of public policies to tackle them should be given priority attention, not just by scholars, but also by politicians and policy-makers.

Title: Modelling dependence between hydrological and meteorological variables measured on several stations Author: Bc. Marie Turčičová Department: Department of Probability and Mathematical Statistics Supervisor: Prof. RNDr. Daniela Jarušková CSc., Czech Technical University in Prague, Faculty of Civil Engineering, Department of Mathematics Abstract: The aim of the thesis is to explore the dependence of daily discharge averages of the Opava river on high daily precipitation values in its basin. Three methods are presented that can be used for analyzing the dependence between high values of random variables. Their application on the studied data is also given. First it is the tail-dependence coefficient that measures the dependence between high values of two continuous random variables. The model for the high quantiles of the discharge at a given precipitation value was first determined non-parametrically by quantile regression and then parametrically through the peaks-over-threshold (POT) method. Keywords: extremal dependence, tail-dependence coefficient, quantile regression, peaks over threshold method

This thesis investigates forecasting performance of Quantile Regression Neural Networks in forecasting multiperiod quantiles of realized volatility and quantiles of returns. It relies on model-free measures of realized variance and its components (realized variance, median realized variance, integrated variance, jump variation and positive and negative semivariances). The data used are S&P 500 futures and WTI Crude Oil futures contracts. Resulting models of returns and volatility have good absolute performance and relative performance in comparison to the linear quantile regression models. In the case of in- sample the models estimated by Quantile Regression Neural Networks provide better estimates than linear quantile regression models and in the case of out-of-sample they are equally good.