Individuals with high anxiety levels from clinical and non-clinical populations tend to exhibit an attentional bias where they selectively allocate more attention to threat stimuli than neutral stimuli, in comparison to individuals with lower anxiety levels. However, longitudinal studies investigating the relations between attentional bias to threat and symptoms of anxiety, depression, and social anxiety––some of the most common mental disorders––are scarce. Using a concurrent and longitudinal design, we investigated the relations between attentional bias to threat and symptoms of anxiety, depression, and social anxiety; concurrently in adulthood (the 30s) as well as longitudinally between young adulthood (the 20s) and adulthood (the 30s). We also investigated whether attentional bias to threat in the 30s moderated and/or mediated the relation between symptoms of psychopathology in the 20s and the same symptoms in the 30s. We found significant concurrent correlations between attentional bias to threat and greater symptoms of anxiety, depression, and social anxiety in the 30s. We also found positive longitudinal correlations between attentional bias to threat in the 30s and symptoms of anxiety (approached significance) and depression (significant) in the 20s. Thus, greater symptoms of internalizing-related psychopathology were associated with greater attentional bias to threat. Attentional bias to threat did not mediate the relation between early psychopathology and later psychopathology, but it did moderate the relation between anxiety in the 20s and social anxiety nearly a decade later. In individuals with greater attentional bias to threat, early anxiety was significantly associated with and predicted greater future social anxiety, but this was not the case for individuals with lower attentional bias to threat. Hence, attentional bias to threat may have a critical role in internalizing-related psychopathology, and interventions targeting it may have preventative and therapeutic potential for mitigating the likelihood of the development and/or persistence of internalizing-related psychopathology. / Thesis / Master of Science (MSc) / People with higher anxiety levels pay more attention to threatening information than neutral information, compared to people with lower anxiety levels. Relatively few studies have investigated the long-term relation between attentional bias to threat and symptoms of mental disorder. Our study investigated the concurrent and longitudinal relations between attentional bias to threat and symptoms of anxiety, depression, and social anxiety. We found that anxiety, depression, and social anxiety in the 30s were concurrently related to greater attentional bias to threat. Additionally, anxiety and depression in the 20s were longitudinally related to greater attentional bias to threat 10 years later. Moreover, people with high anxiety and high attentional bias to threat were more likely to experience social anxiety in the future than people with high anxiety but low attentional bias to threat. Therefore, attentional bias to threat might have a critical role in the development and/or persistence of some mental disorders.

The introduction of Science Education Standards for the K-12 education system of the United States and those of Bhutan is aimed at providing students with equitable access to quality science instruction and to promote a scientifically literate society. These reforms in education systems require teachers who are well prepared to translate the reform documents, (i.e., Science Standards and curriculum goals) into effective daily classroom instruction. Reform efforts by the states or the federal government involve significant investment of financial and other resources. But all too often the teachers are not prepared to implement reforms in their classroom professional practices due to lack of proper (and timely) professional development and necessary support. To design professional development and educative materials to enable teachers’ implementation of these reform ideas, we need research into the current practices and challenges in implementing reform ideas by pre-service science teachers and in-service science teachers. Science education programs and research has increasingly placed emphasis on the need for sound theoretical models to support educational program development and implementation. This is a case study based on a multi-compartment network model designed to examine aspects of current practices in implementing reform ideas, and furthermore to identify areas of best practices and areas of improvement (e.g., professional practices and interpretation of theories of teaching and learning). The core of the model is grounded in the principles of social constructivism, and the relevant theories and practices guiding teacher education forms the multi-compartments of a network box model that focuses on how the components interact with each other in a teacher education program. It is intended to provide a holistic picture of how the teacher education practices and reform implementation goals interact. This model was initially applied in an investigation of practices of a cohort of pre-service science teachers in a teacher education program in the U.S. This constituted phase one of this research. Subsequently, it also was applied to a pre-service science teacher education in Bhutan. There, the course involved translation of reform ideas into the pre-service teacher’s professional practices, in this latter case science Standards (referred to as Bhutan Goals of Science Education) in Bhutan. In parallel to the U.S.-based study, a cohort of pre-service science teachers in a science education course and relevant expert science teacher educators participated in the Bhutan study. The study was carried out in two separate case studies with distinct contextual characteristics. The U.S. case study part one was designed in some ways to pilot the analytical model. The Bhutan case study part two is an extension of the first study. In both the U.S. and Bhutan study, the pre-service science teachers and expert science teacher educators showed strong degree of coherence in terms of their ranking of the science Standards and their rationale behind the ranking, indicating a certain degree of evidence for a community of practice. However, further inquiry revealed that pre-service science teachers struggled to effectively incorporate Standards in their lesson plans. Furthermore, the findings from the two-part study provide some insights into how this analytical model can be applied to science teacher education engaged in reform implementation, across different institutional and cultural contexts. Both quantitative and qualitative evidence were obtained and analyzed using descriptive statistical analysis methods and qualitative data analysis techniques.

Challenges of financial forecasting, such as a dearth of independent samples and non- stationary underlying process, limit the relevance of conventional machine learning towards financial forecasting. Meta-learning approaches alleviate some of these is- sues by allowing the model to generalize across unrelated or loosely related tasks with few observations per task. The neural process family achieves this by con- ditioning forecasts based on a supplied context set at test time. Despite promise, meta-learning approaches remain underutilized in finance. To our knowledge, ours is the first application of neural processes to realized volatility (RV) forecasting and financial forecasting in general. We propose a hybrid temporal convolutional network attentive neural process (ANP- TCN) for the purpose of financial forecasting. The ANP-TCN combines a conven- tional and performant financial time series embedding model (TCN) with an ANP objective. We found ANP-TCN variant models outperformed the base TCN for equity index realized volatility forecasting. In addition, when stack-ensembled with a tree- based model to forecast a trading signal, the ANP-TCN outperformed the baseline buy-and-hold strategy and base TCN model in out-of-sample performance. Across four liquid US equity indices (incl. S&P 500) tested over ∼15 years, the best long-short models (reported by median trajectory) resulted in the following out-of-sample (∼3 years) performance ranges: directional accuracy of 58.65% to 62.26%, compound an- nual growth rate (CAGR) of 0.2176 to 0.4534, and annualized Sharpe ratio of 2.1564 to 3.3375. All project code can be found at: https://github.com/kpa28-git/thesis-code.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.