This thesis focuses on the development of semiparametric estimation methods for a class of time series models using shape constraints. Many of the existing time series models assume the noise follows some known parametric distributions. Typical examples are the Gaussian and t distributions. Then the model parameters are estimated by maximizing the resultant likelihood function. As an example, the autoregressive moving average (ARMA) models (Brockwell and Davis, 2009) assume Gaussian noise sequence and are estimated under the causal-invertible constraint by maximizing the Gaussian likelihood. Although the same estimates can also be used in the causal-invertible non-Gaussian case, they are not asymptotically optimal (Rosenblatt, 2012). Moreover, for the noncausal/noninvertible cases, the Gaussian likelihood estimation procedure is not applicable, since any second-order based methods cannot distinguish between causal-invertible and noncausal/noninvertible models (Brockwell and Davis,2009). As a result, many estimation methods for noncausal/noninvertible ARMA models assume the noise follows a known non-Gaussian distribution, like a Laplace distribution or a t distribution. To relax this distributional assumption and allow noncausal/noninvertible models, we borrow ideas from nonparametric shape-constraint density estimation and propose a semiparametric estimation procedure for general ARMA models by projecting the underlying noise distribution onto the space of log-concave measures (Cule and Samworth, 2010; Dümbgen et al., 2011). We show the maximum likelihood estimators in this semiparametric setting are consistent. In fact, the MLE is robust to the misspecification of log-concavity in cases where the true distribution of the noise is close to its log-concave projection. We derive a lower bound for the best asymptotic variance of regular estimators at rate sqrt(n) for AR models and construct a semiparametric efficient estimator. We also consider modeling time series of counts with shape constraints. Many of the formulated models for count time series are expressed via a pair of generalized state-space equations. In this set-up, the observation equation specifies the conditional distribution of the observation Yt at time t given a state-variable Xt. For count time series, this conditional distribution is usually specified as coming from a known parametric family such as the Poisson or the Negative Binomial distribution. To relax this formal parametric framework, we introduce a concave shape constraint into the one-parameter exponential family. This essentially amounts to assuming that the reference measure is log-concave. In this fashion, we are able to extend the class of observation-driven models studied in Davis and Liu (2016). Under this formulation, there exists a stationary and ergodic solution to the state-space model. In this new modeling framework, we consider the inference problem of estimating both the parameters of the mean model and the log-concave function, corresponding to the reference measure. We then compute and maximize the likelihood function over both the parameters associated with the mean function and the reference measure subject to a concavity constraint. The estimator of the mean function and the conditional distribution are shown to be consistent and perform well compared to a full parametric model specification. The finite sample behavior of the estimators are studied via simulation and two empirical examples are provided to illustrate the methodology.

No description available.

Dans cette thèse, on s'intéresse au développement de nouvelles méthodes d'auscultation GPR pour déterminer la géométrie et la structure des chaussées. Cette thèse a deux objectifs principaux. Tout d'abord, elle a pour but d'améliorer la compréhension des mécanismes de diffusion à très large bande dans un milieu stratifié composé d'interfaces rugueuses. Avec l'augmentation des fréquences d'utilisation de différents systèmes, les interfaces de chaussée ne peuvent plus être considérées comme planes. Ainsi, la rugosité des interfaces doit être prise en compte dans la modélisation de la propagation. Donc, une analyse de l'influence de cette rugosité sur l'onde rétrodiffusée a été réalisée. Elle a permis de montrer que la rugosité induit une décroissance en fréquence de l'amplitude des échos. Cette décroissance a ensuite été introduite dans le modèle du signal. Dans un second temps, plusieurs méthodes de traitement de signal ont été proposées pour estimer conjointement les paramètres de rugosité et d'épaisseur. D'abord, des méthodes multidimensionnelles ont été proposées en prenant en compte l'influence de la rugosité. Ensuite, afin de réduire la charge de calcul, des méthodes monodimensionnelles ont été proposées. Ces méthodes ont été évaluées à partir de signaux simulés. Les résultats ont montré de bonnes performances pour l'estimation des temps de retard et des paramètres de rugosité des interfaces. Enfin, les méthodes de traitement proposées dans ce manuscrit ont été testées sur des données expérimentales, qui permettent de valider les résultats théoriques et de montrer la faisabilité de la mesure de couches minces de chaussée et du paramètre de rugosité. / In this thesis, we focus on the development of new GPR methods to estimate the pavement structure. This thesis has two main objectives. First, it aims to improve the understanding of the scattering mechanisms for large-band radars in a stratified medium composed of rough interfaces. With increasing frequencies, pavement interfaces can no longer be considered as flat. The interface roughness must be taken into account in the propagation modelling. Thus, the influence of the roughness has been analysed. It has been shown that the interface roughness provides a continuous frequency decay of the magnitude of the echoes. This continuous frequency decay has then been introduced into the signal model. Secondly, several signal processing methods have been proposed to jointly estimate the roughness and thickness of pavement. Thus, multidimensional methods have been proposed by taking into account the roughness.Then, in order to reduce the computational burden, one-dimensional methods have also been proposed. From simulations, it can be seen that the proposed algorithms provide a good performance in parameter estimations (time delay, permittivity, roughness and thickness). Finally, the proposed signal processing methods are tested on experimental data. The results confirm the theoretical prediction. They show the feasibility to estimate both the thickness of thin pavements and roughness parameter.

The present work deals with selected methods suitable to work with financial time series. Firstly, univariate linear models ARMA are introduced, followed by the description of volatility models ARCH and their generalization to GARCH models. There are many modifications of standard GARCH models designed with respect to the nature of financial data, some of which are presented. Another part of the work dealing with multiple time series focuses on VAR models and bivariate GARCH models. The most important part of the work are practical examples of building the theoretically described models in various types of software with built-in procedures for time series analysis. We apply five different types of commercial and non-commercial software, namely EViews, Mathematica, R, S-PLUS and XploRe. The used software products are presented and compared in terms of their capabilities and the results obtained for particular methods.

The Real-time Specification of Java (RTSJ) is an intention to introduce Java as a language for developing real-time system. However, the complexity of their development and a non-trivial programming model of RTSJ with its manual memory management often lead to programming errors. To mitigate the development of RTSJ systems it would be beneficial to provide an internal domain-specific language (DSL) extending the Java language which would allow to develop the systems in more intuitive and safer way. However, it is needed to find compromise between solution's power and level of usability, because this two attributes go often against each other. One possible way of DSLs creation concerns the Meta-Programming System (MPS). It allows to develop new domain-specific languages and corresponding projectional editors enabling different views on code. This thesis proposes a design and implementation of the DSL on the top of the MPS platform and corresponding code generator enabling development of RTSJ systems. Furthermore, the thesis provides a simple case-study to evaluate a proposed DSL. Additionally, the thesis assesses the suitability of MPS as a DSL-development platform.

This thesis seeks to examine time in the secondary world fantasies of Patricia A. McKillip. Very little work has been done previously on McKillip, and none examines such a broad range of her works. Taking a strict definition of secondary world, I investigate McKillip’s fantasy books that fit within this parameter to see whether there is a unifying principal behind time, in all its forms, in her work. Although time has been examined in fantasies that are obviously about time, very little has been done in the style of Mark Currie or Paul Riceour, who examine time in books that contain time but do not seem to be about time. I investigate time in terms of an overall chronotope, and argue that this seems to be one of the past. I argue that McKillip’s works and other fantasy books like hers have a grammar of the past, and that everything in their works is influenced, to a degree, by this grammar. Thematically organised chapters examine sixteen of McKillip’s immersive fantasies. The thesis begins with an investigation of the overall chronotope of McKillip’s books and the influence this has on her works. It then examines “active time”: time which is used in an active way to undo and heal wrongs of the past. McKillip’s use of legends to add depth and age to her stories is explored. Her pastoral works, those with a nostalgic connection with nature, are examined. The sometimes counterpoint of the pastoral, cities, are then investigated and found to be places of influenced by time passing in the form of age and political era. Lastly, several of McKillip’s characters are examined to show how time has affected them and affected their interactions with those around them.

O tênis de mesa é um dos esportes de raquete mais rápidos do mundo, exigindo muito pouco tempo para planejar e executar seus movimentos. Neste sentido, detectar a quantidade de rotação da bola, velocidade e direção torna-se fundamental para produzir respostas motoras eficientes. Entretanto, existem algumas questões que ainda não foram exploradas como o quanto a disponibilidade das informações auditiva e visual interfere na acurácia da percepção da rotação da bola e no tempo necessário para perceber estas rotações. O propósito do presente estudo foi analisar a influência da informação auditiva e visual na percepção da rotação da bola e no desempenho no tênis de mesa. Os atletas profissionais foram submetidos a dois estudos: o primeiro destinado a verificar a contribuição da informação auditiva e visual na percepção da rotação da bola e no tempo necessário para essa percepção. O segundo, proposto para avaliar a influência da informação auditiva no desempenho da tarefa de contra-ataque no tênis de mesa. Participaram do estudo 22 mesatenistas de alto rendimento de ambos os sexos (21 homens e 1 mulher) com média de idade de 22,55 ± 6,15. O teste de percepção de rotação (estudo 1) analisou o número de acertos e o tempo de reação por meio de uma rotina desenvolvida no software Matlab (Apêndice D). Os atletas visualizavam a tela de um computador e respondiam em um teclado o mais rápido possível. Os atletas analisaram um movimento de forehand nas intensidades de rotação: muita rotação (bola girando a aproximadamente 140 r/s); média (bola girando a aproximadamente 105 r/s); pouca rotação (bola girando a aproximadamente 84 r/s); batida chapada (bola girando a aproximadamente 70 r/s), nas seguintes condições: a) informação auditiva; b) informação visual; c) informação auditiva e visual. Os atletas analisaram 20 bolas para cada condição sendo cinco de cada tipo de efeito. Para o teste de desempenho (estudo 2) o atleta teve que responder adequadamente com o movimento de contra-ataque de forehand em um alvo previamente marcado na mesa. Foram golpeadas 80 bolas nas mesmas intensidades de rotação mencionadas acima. Os atletas foram avaliados em duas condições: com informação auditiva e sem informação auditiva. Os resultados do estudo 1 revelaram que mesatenistas foram capazes de perceber a magnitude de rotação da bola nas três condições. O percentual de acertos foi de 77,5%, 91,14% e 96,36% para as condições de informação auditiva, visual e combinadas, respectivamente. Quanto ao tempo de reação, os resultados revelaram que na condição somente informação auditiva (2,32 ± 0,41 s) os atletas demoraram mais tempo para responder do que em relação a condição somente informação visual (2,07 ± 0,36 s), já na condição informação auditiva e visual (1,85 ± 0,35 s) os atletas se mostraram mais rápidos ainda concordando com as informações número de acertos obtidas neste estudo. Para o estudo 2 os resultados revelaram que quando os atletas contra-atacaram na condição informação auditiva, o número de acertos foi de 37,69 ± 5,43 enquanto, quando a audição dos atletas foi privada a média de acertos foi 33,15 ± 6,84. Os resultados apontaram que mesatenistas foram capazes de utilizar as informações auditiva e visual para perceber a magnitude da rotação da bola no golpe forehand, porém, no estudo 1 quando as informações foram combinadas o número de acertos aumentou 5,22% e o tempo de reação diminuiu 0,22 s. No estudo 2 Quando a informação auditiva foi privada houve uma diminuição na média do número de acertos de 4,54. / Table tennis is one of the fastest racket sports in the world, requiring very little time to plan and execute your moves. In this sense, detecting a quantity of ball spin, velocity and direction becomes fundamental to produce efficient motor responses. However, there are some questions that have not yet been explored for example how much the availability of auditory and visual information interferes with the accuracy of the perception of the rotation of the ball and the time needed to perceive these rotations. The purpose of the present study is to analyze the influence of auditory and visual information on the perception of ball rotation and performance in table tennis. The volunteers were submitted to two studies: the first one to verify the contribution of the auditory and visual information in the perception of the rotation of the ball and in the time necessary for this perception. The second, proposed to evaluate the influence of auditory information on the performance of the counter-attack task in table tennis. Twenty-two high-level table tennis players men and women (21 men and 1 woman) with a mean age of 22.55 ± 6.15 participated in the study. The rotation perception test (study 1) analyzed the number of hits and the reaction time through a routine developed in Matlab software. The athletes visualized the screen of a computer and answered in a keyboard as soon as possible. The athletes analyzed a forehand movement in the rotation intensities: fast spin (spinning ball foward at 140 r/s); medium spin (105 r/s); slow spin (84 r/s); flat hit (70 r/s), under the following conditions: a) auditory information; b) visual information; c) auditory and visual information. The athletes analyzed 20 balls for each condition, five of each type of rotation. For the performance test (study 2) the athlete had to respond adequately with the forehand counterattack movement on a previously marked target at the table. 80 balls were struck at the same intensities of rotation mentioned above. The athletes were evaluated in two conditions: with auditory information and without auditory information. The results of the study 1 revealed that table tennis players were able to perceive the magnitude of rotation of the ball in the three conditions. The percentage of correct answers was 77.5%, 91.14% and 96.36% for auditory, visual and combined information conditions, respectively. Regarding the reaction time, the results revealed that in the condition only auditory information (2.32 ± 0.41 s) the athletes took more time to respond, since in the condition only visual information (2.07 ± 0.36 s) delayed a little less and in the condition auditory and visual information (1.85 ± 0.35 s) the athletes showed themselves to be faster still agreeing with the information number of hits obtained in this study. For the study 2 the results revealed that when the athletes counterattacked in the auditory information condition, the number of hits was 37.69 ± 5.43, whereas when the hearing of the athletes was private the average hit was 33.15 ± 6.84. The results indicated that table tennis players were able to use auditory and visual information to realize the magnitude of the ball rotation in the forehand blow, however, in study 1 when the informations were combined the number of right answered increased 5.22% and the reaction time decreased by 0.22 s. In study 2 When the auditory information was depriving, there was a decrease in the average number of hits 4.54.

The question how much internationalization is beneficial for emerging-market small and medium enterprises (EM SMEs) remains challenging to answer for both international business (IB) scholars and managers. We explore export strategies of first time exporters and focus on the scope of EM SMEs internationalization activities. We tackle the question whether more focused or more diversified internationalization through exporting is beneficial for EM SMEs. We examine the impact of foreign market (geographic) diversification, product diversification and export intensity on firm performance of an entire population of EM SMEs from an emerging east European market. In addition, we test whether a complex export strategy-an export strategy of simultaneous product- and geographic export diversification-is beneficial for EM SMEs. We use a panel population data of first time Slovenian exporters in the period 1994-2012. We find that diversified internationalization, both in terms of product and foreign market diversity, significantly improves productivity and sales performance for EM SMEs. Furthermore, EM SMEs with complex export strategies enjoy significantly improved productivity and sales performance. / Series: Working Papers / Institute for International Business

This thesis describes work on a novel 3D imaging system that successfully implements optical feedback and noise rejection mechanisms. The system is a combination of three relatively new technologies, namely, holographic projection, Time of Flight (ToF) ranging and Single Photon Avalanche Diode (SPAD) sensors. Holographic projection is used to provide structured illumination with optical feedback instead of more commonly used uniform illumination in similar imaging systems. It is obtained using a Ferro-electric Liquid Crystal on Silicon Spatial Light Modulator (FLCoS SLM). The structured illumination with optical feedback can be operated at up to 60 Hz with the current device, and has been shown to provide an average gain of about 1.56 in useful light levels. Alternatively, a gain over a limited area of up to a factor of 9 is possible with the current system. Time of Flight ranging is a method of choice for the system when depth estimation is concerned. It works even at very low light levels and allows for sub-centimetre depth resolution. ToF method was implemented using 20 MHz laser diode with 50 ps pulse duration and 200 mW peak power, as well as a SPAD sensor. The SPAD sensor consisted of a 32 32 array of 50 μm pixels, each with 10 bit Time to Digital Converter (TDC) with 50 ps timing resolution. Sensor pixels feature 100 Hz mean Dark Count Rate (DCR). The use of SPAD sensors with an adaptive sensing algorithm presented in this work has been demonstrated to reduce effective noise levels as seen by the sensor by a factor of 16. As a result, a significant gain in depth resolution can be achieved. The quantification of this gain is explained in more detail within this work. Furthermore, the work describes in detail system design, methodology of experimental procedure as well as different algorithms essential to the correct operation of the system. Significant amount of time is dedicated to diffraction pattern generation for the use in holographic projection, as well as modelling of photon detection in SPAD sensors and associated peak detection necessary to extract depth information from histograms of timed of photons. Moreover, the thesis discusses potential applications for the system based on the results of system characterisation presented in this work. The current state of the system suggests best suitability for gaming and machine vision applications. Finally, the work offers potential solutions to the practical issues that remain unresolved in the current system, alternatives for components used and paths for potential future development of the system proposed.

The fundamental objective of the analysis of financial time series is to unveil the random mechanism, i.e. the probability law, underlying financial data. The effort to identify the truth that governs the observations involves proposing and estimating reasonable statistical models that well explain the empirical features of data. This thesis develops some new nonparametric tools that can be exploited in this context; the efficacy and validity of their use are supported by computational advancements and surging availability of large/complex (`big') data sets. Chapter 1 investigates the conditional first moment properties of financial returns. We propose multivariate extensions of the popular Variance Ratio (VR) statistic, aiming to test linear predictability of returns and weak-form market efficiency. We construct asymptotic distribution theories for the statistics and scalar functions thereof under the null hypothesis of no predictability. The imposed assumptions are weaker than those widely adopted in the literature, and in our view more credible with regard to the underlying data generating process we expect for stock returns. It is also shown that the limit theories can be extended to the long horizon and large dimension cases, and also to allow for a time varying risk premium. Our methods are applied to CRSP weekly returns from 1962 to 2013; the joint tests of the multivariate hypothesis reject the null at the 1% level for all horizons considered. Chapter 2 is about nonparametric estimation of conditional moments. We propose a local constant type estimator that operates with an infinite number of conditioning variables; this enables a direct estimation of many objects of econometric interest that have dependence upon the infinite past. We show pointwise and uniform consistency of the estimator and establish its asymptotic nomality in various static and dynamic regressions context. The optimal rate of estimation turns out to be of logarithmic order, and the precise rate depends on the Lambert W function, the smoothness of the regression operator and the dependence of the data in a non-trivial way. The theories are applied to investigate the intertemporal risk-return relation for the aggregate stock market. We report an overall positive risk-return relation on the S&P 500 daily data from 1950-2017, and find evidence of strong time variation and counter-cyclical behaviour in risk aversion. Lastly, Chapter 3 concerns nonparametric volatility estimation with high frequency time series. While data observed at finer time scale than daily provide rich information, their distinctive empirical properties bring new challenges in their analysis. We propose a Fourier domain based estimator for multivariate ex-post volatility that is robust to two major hurdles in high frequency finance: asynchronicity in observations and the presence of microstructure noise. Asymptotic properties are derived under some mild conditions. Simulation studies show our method outperforms time domain estimators when two assets with different liquidity are traded asynchronously.