In single-case research, multiple-baseline (MB) design is the most widely used design in practical settings. It provides the opportunity to estimate the treatment effect based on not only within-series comparisons of treatment phase to baseline phase observations, but also time-specific between-series comparisons of observations from those that have started treatment to those that are still in the baseline. In MB studies, the average treatment effect and the variation of these effects across multiple participants can be estimated using various statistical modeling methods. Recently, two types of statistical modeling methods were proposed for analyzing MB studies: a) within-series model and b) between-series model. The within-series model is a typical two-level multilevel modeling approach analyzing the measurement occasions within a participant, whereas the between-series model is an alternative modeling approach analyzing participants’ measurement occasions at certain time points, where some participants are in the baseline phase and others are in the treatment phase. Parameters of both within- and between-series models are generally estimated with restricted maximum likelihood (ReML) estimation and ReML is developed based on the assumption of normality (Hox, et al., 2010; Raudenbush & Bryk, 2002). However, in practical educational and psychological settings, observed data may not be easily assumed to be normal. Therefore, the purpose of this study is to investigate the robustness of analyzing MB studies with the within- and between-series models when level-1 errors are non-normal. A Monte Carlo study was conducted under the conditions where level-1 errors were generated from non-normal distributions in which skewness and kurtosis of the distribution were manipulated. Four statistical approaches were considered for comparison based on theoretical and/or empirical rationales. The approaches were defined by the crossing of two analytic decisions: a) whether to use a within- or between-series estimate of effect and b) whether to use REML estimation with Kenward-Roger adjustment for inferences or Bayesian estimation and inference. The accuracy of parameter estimation and statistical power and Type I error were systematically analyzed. The results of the study showed the within- and between-series models are robust to the non-normality of the level-1 error variance. Both within- and between-series models estimated the treatment effect accurately and statistical inferences were acceptable. ReML and Bayesian estimations also showed similar results in the current study. Applications and implications for applied and methodology researchers are discussed based on the findings of the study.

Accurate prediction and control pervades domains such as engineering, physics, chemistry, and biology. Often, it is discovered that the systems under consideration cannot be well represented by linear, periodic nor random data. It has been shown that these systems exhibit deterministic chaos behavior. Deterministic chaos describes systems which are governed by deterministic rules but whose data appear to be random or quasi-periodic distributions. Deterministically chaotic systems characteristically exhibit sensitive dependence upon initial conditions manifested through rapid divergence of states initially close to one another. Due to this characterization, it has been deemed impossible to accurately predict future states of these systems for longer time scales. Fortunately, the deterministic nature of these systems allows for accurate short term predictions, given the dynamics of the system are well understood. This fact has been exploited in the research community and has resulted in various algorithms for short term predictions. Detection of normality in deterministically chaotic systems is critical in understanding the system sufficiently to able to predict future states. Due to the sensitivity to initial conditions, the detection of normal operational states for a deterministically chaotic system can be challenging. The addition of small perturbations to the system, which may result in bifurcation of the normal states, further complicates the problem. The detection of anomalies and prediction of future states of the chaotic system allows for greater understanding of these systems. The goal of this research is to produce methodologies for determining states of normality for deterministically chaotic systems, detection of anomalous behavior, and the more accurate prediction of future states of the system. Additionally, the ability to detect subtle system state changes is discussed. The dissertation addresses these goals by proposing new representational techniques and novel prediction methodologies. The value and efficiency of these methods are explored in various case studies. Presented is an overview of chaotic systems with examples taken from the real world. A representation schema for rapid understanding of the various states of deterministically chaotic systems is presented. This schema is then used to detect anomalies and system state changes. Additionally, a novel prediction methodology which utilizes Lyapunov exponents to facilitate longer term prediction accuracy is presented and compared with other nonlinear prediction methodologies. These novel methodologies are then demonstrated on applications such as wind energy, cyber security and classification of social networks.

In my final work I deal with the topic of handicap, disability and otherness. Any otherness causes fear in society, fear of ignorance. Being healthy, "normal," means the assumption of a full-fledged life of modern man. A disabled person, whether physically or mentally, does not meet these priorities and ideas of normality. I try to point out the handicap positively, not to hide it. I also work on the qualities of beauty and ugliness because they are very similar to health and disability issues. My final work is focused on the affected bodies. The result is a set of exposed objects that, in conjunction with the installation, can act as luxury goods. The work consists of about fifteen hand-sewn gloves of various shapes and materials. They are very extraordinary objects that only "sit" for the chosen. The aim of the work is to contribute to works that seek to promote disability in art or culture.

In the era of Internet of Things (IoT) and big data, collecting, processing and analyzing enormous data faces unprecedented challenges even when being stored in preprocessed form. Anomaly detection, statistically viewed as identifying outliers having low probabilities from the modelling of data distribution p(x), becomes more crucial. In this Master thesis, two (supervised and unsupervised) novel deep anomaly detection frameworks are presented which can achieve state-of-art performance on a range of datasets. Capsule net is an advanced artificial neural network, being able to encode intrinsic spatial relationship between parts and a whole. This property allows it to work as both a classifier and a deep autoencoder. Taking this advantage of CapsNet, a new anomaly detection technique named AnoCapsNet is proposed and three normality score functions are designed: prediction-probability-based (PP-based) normality score function, reconstruction-error-based (RE-based) normality score function, and a normality score function that combines prediction-probability-based and reconstruction-error-based together (named as PP+RE-based normality score function) for evaluating the "outlierness" of unseen images. The results on four datasets demonstrate that the PP-based method performs consistently well, while the RE-based approach is relatively sensitive to the similarity between labeled and unlabeled images. The PP+RE-based approach effectively takes advantages of both methods and achieves state-of-the-art results. In many situations, neither the domain of anomalous samples can be fully understood, nor the domain of the normal samples is straightforward. Thus deep generative models are more suitable than supervised methods in such cases. As a variant of variational autoencoder (VAE), beta-VAE is designed for automated discovery of interpretable factorised latent representations from raw image data in a completely unsupervised manner. The t-Distributed Stochastic Neighbor Embedding (t-SNE), an unsupervised non-linear technique primarily used for data exploration and visualizing high-dimensional data, has advantages at creating a single map that reveals local and important global structure at many different scales. Taking advantages of both disentangled representation learning (using beta-VAE as an implementation) and low-dimensional neighbor embedding (using t-SNE as an implementation), another novel anomaly detection approach named AnoDM (stands for Anomaly detection based on unsupervised Disentangled representation learning and Manifold learning) is presented. A new anomaly score function is defined by combining (1) beta-VAE's reconstruction error, and (2) latent representations' distances in the t-SNE space. This is a general framework, thus any disentangled representation learning and low-dimensional embedding techniques can be applied. AnoDM is evaluated on both image and time-series data and achieves better results than models that use just one of the two measures and other existing advanced deep learning methods.

This paper examines whether the CPI and real GDP for the US exhibit nonlinear reversion to trend as recently concluded by Beechey and Österholm [Beechey, M. and Österholm, P., 2008. Revisiting the uncertain unit root in GDP and CPI: testing for nonlinear trend reversion. Economics Letters 100, 221-223]. The wild bootstrap is used to correct for non-normality and heteroscedasticity in a nonlinear unit root test. The use of 'wild bootstrapped' critical values affects test conclusions in some cases. Results also are sensitive to the sample period examined.

Kurtosis can be measured in more than one way. A modification of Geary's measure of kurtosis is shown to be more sensitive to kurtosis in the center of the distribution while Pearson's measure of kurtosis is more sensitive to kurtosis in the tails of the distribution. The modified Geary measure and the Pearson measure are used to define a joint test of kurtosis that has high uniform power across a very wide range of symmetric nonnormal distributions.

Material failure plays an important role in human life. By investigating the failure mechanisms, people can more precisely predict the failure conditions to develop new products, to enhance product performances, and most importantly, to save lives. This work consists of three parts corresponding to three different failure mechanisms in metals, i.e., the localized necking in sheet metals, the bifurcation in bulk and sheet metals, and the ductile fracture induced by the void nucleation, growth, and coalescence. The objective of the first part is to model the localized necking in anisotropic sheet metals to demonstrate that localized geometric softening at a certain stage of deformation rather than the initial defects is the main cause of localized necking. The sheet is assumed to have no initial geometric defects. The deformation process is divided into two stages. The critical strains for a neck to form are obtained from a Considère-type criterion. The defect ratio at the neck formation is obtained using an energy-based approach. The neck evolution is considered. A novel failure criterion is proposed. Two types of necks are fond to be most competitive to cause material failure during continued deformation. The forming limit curves are hereby found to exhibit different characteristics in different region. The predicted forming limit curve for 2036-T4 aluminum is found to fit with the experimental results well. The sheet thickness, the strain hardening behavior, and plastic anisotropy are found to affect the sheet metal formability. More realistic yield criterions and strain hardening behaviors can be implemented into the proposed model. This part provides an alternative approach to modeling the localized necking in anisotropic sheet metals. The objective of the second part is to model the bifurcation in anisotropic bulk and sheet metals to couple plastic anisotropy and the strain hardening/softening behavior and also to identify different bifurcation modes in sheet metals. The material is assumed to exhibit a non-linear strain hardening/softening behavior and to obey the Hill-type Drucker-Prager yield criterion along with a non associated flow rule. The constitutive relations and the conditions for bifurcation in bulk and sheet metals are derived. The internal friction coefficient, plastic anisotropy, the terms introduced by the co-rotational stress rates, and the terms introduced by the stress resultant equilibrium are found to affect the onset of bifurcation. Two bifurcation modes are found to exist in sheet metals. More realistic material properties can be implemented into the proposed model. This part provides an applicable approach to modeling the bifurcation in anisotropic bulk and sheet metals. The objective of the third part is to derive the constitutive relations for porous metals using generalized Green’s functions to better understand the micromechanism of the ductile fracture in metals. The porous metals are assumed to consist of an isotropic, rigid-perfectly plastic matrix and numerous periodically distributed voids and to be subject to non-equal biaxial or triaxial extension. Two types of hollow cuboid RVEs are employed represent the typical properties of porous metals with cylindrical and spherical voids. The microscopic velocity fields are obtained using generalized Green’s functions. The constitutive relations are derived using the kinematic approach of the Hill-Mandel homogenization theory and the limit analysis theory. The macroscopic mean stress, the porosity, the unperturbed velocity field, and the void distribution anisotropy are found to affect the macroscopic effective stress and the microscopic effective rate of deformation field. The proposed model is found to provide a rigorous upper bound. More complicated matrix properties (e.g., plastic anisotropy) and void shapes can be implemented into the proposed model. This part provides an alternative approach to deriving the constitutive relations for porous metals.

The main purpose of this dissertation is to investigate directional constraint qualifications and necessary optimality conditions for nonsmooth set-constrained mathematical programs. First, we study sufficient conditions for metric subregularity of the set-constrained system. We introduce the directional version of the quasi-/pseudo-normality as a sufficient condition for metric subregularity, which is weaker than the classical quasi-/pseudo-normality, respectively. Then we apply our results to complementarity and Karush-Kuhn-Tucker systems. Secondly, we study directional optimality conditions of bilevel programs. It is well-known that the value function reformulation of bilevel programs provides equivalent single-level optimization problems， which are nonsmooth and never satisfy the usual constraint qualifications such as the Mangasarian-Fromovitz constraint qualification (MFCQ). We show that even the first-order sufficient condition for metric subregularity (which is generally weaker than MFCQ) fails at each feasible point of bilevel programs. We introduce the directional Clarke calmness condition and show that under the directional Clarke calmness condition, the directional necessary optimality condition holds. We perform directional sensitivity analysis of the value function and propose the directional quasi-normality as a sufficient condition for the directional Clarke calmness. / Graduate / 2021-07-07

Vi har genomfört en kvalitativ intervjustudie om föräldrars erfarenheter av en vardag tillsammans med ett barn som har Downs syndrom, i synnerhet när det gäller barnets ungdomstid. Den insamlade empirin har tolkats med hjälp av Beckers (2006) teori om avvikelse, Goffmans (2001) teori om stigma samt Goffmans (2009) teori om interaktion i det vardagliga sociala livet. Föräldern ingår i ett allmänt system av normalitet, såsom övriga samhällsmedlemmar. Studiens fokus ligger på förälderns agerande utefter detta i förhållande till den situation som barnets funktionsnedsättning för med sig. Downs syndrom medför en utvecklingsstörning, vilket innebär att barnets kroppsliga och mentala utveckling inte alltid är i fas med varandra. En följd av detta är att småbarns- och ungdomstiden förlängs, vilket föräldrarna i vår studie upplever som en svårighet. Svårigheten ligger bland annat i att barnets självständighetsutveckling skiljer sig från det som i allmänhet anses vara normalt och därmed skiljer sig även förälderns roll i denna utveckling. Studien visar att föreställningar om normalitet ständigt är närvarande i föräldrarnas berättelser. / We have made a qualitative interview study about parents' experiences of every day life with a child who has Down syndrome, particularly with regard to the child's youth. The empirical data collected has been interpreted using Becker's (2006) theory of deviance, Goffman's (2001) theory of stigma and Goffman's (2009) theory of interaction in everyday social life. The parent are included in a general system of normality, just as any other member of society. The focus of the study is parent's acting in relation to normality and to the situation that the child's disability causes. Down syndrome results in a development disorder, which means that child's physical and mental development not always is in phase with each other. This causes an extension of the childhood and youth, which the parents in our study perceive as difficult. The difficulty lies among other things in that the child's development of independence differs from what is generally considered to be normal and that the parent's role in this development consequently also differs. The study shows that ideas of normality are constantly present in the parents' narratives.

Goodness-of-fit and correlation tests are considered for dependent univariate data that arises when multivariate data is projected to the real line with a data-suggested linear transformation. Specifically, tests for multivariate normality are investigated. Let { } i Y be a sequence of independent k-variate normal random vectors, and let 0 d be a fixed linear transform from Rk to R . For a sequence of linear transforms { ( )} 1 , , n d Y Y converging almost surely to 0 d , the weak convergence of the empirical process of the standardized projections from d to a tight Gaussian process is established. This tight Gaussian process is identical to that which arises in the univariate case where the mean and standard deviation are estimated by the sample mean and sample standard deviation (Wood, 1975). The tight Gaussian process determines the limiting null distribution of E.D.F. goodness-of-fit statistics applied to the process of the projections. A class of tests for multivariate normality, which are based on the Shapiro-Wilk statistic and the related correlation statistics applied to the dependent univariate data that arises with a data-suggested linear transformation, is also considered. The asymptotic properties for these statistics are established. In both cases, the statistics based on random linear transformations are shown to be asymptotically equivalent to the statistics using the fixed linear transformation. The statistics based on the fixed linear transformation have same critical points as the corresponding tests of univariate normality; this allows an easy implementation of these tests for multivariate normality. Of particular interest are two classes of transforms that have been previously considered for testing multivariate normality and are special cases of the projections considered here. The first transformation, originally considered by Wood (1981), is based on a symmetric decomposition of the inverse sample covariance matrix. The asymptotic properties of these transformed empirical processes were fully developed using classical results. The second class of transforms is the principal components that arise in principal component analysis. Peterson and Stromberg (1998) suggested using these transforms with the univariate Shapiro-Wilk statistic. Using these suggested projections, the limiting distribution of the E.D.F. goodness-of-fit and correlation statistics are developed.