In the first essay, I document sample-specific and time period-specific style returns in two distinct sets of U.S. equities: Fama and French style portfolios and the S&P 1500 style indexes. The value and size effects are apparent in the Fama and French portfolios. Only the size effect is evident in the S&P data. In general, the Fama and French style returns are greater than those of the S&P 1500 styles. Style returns tend to be time-varying and exhibit momentum over a variety of formation period-holding period horizons. In the second essay, I utilize a bootstrap procedure to test for the presence of styleswitchers - as defined in Barberis and Shleifer (2003). I document style winner and loser continuations. There are some periods when no matter which particular style won (lost) in the past, it is more likely to continue winning (losing) in the future. I also test some Barberis and Shleifer (2003) propositions regarding style momentum. One proposition holds that Sharpe ratios from style-level momentum strategies should be at least as large as asset-level momentum Sharpe ratios. While many style momentum strategies generate significant returns, the implied Sharpe ratios are lower than those reported for asset-level momentum strategies. The Barberis and Shleifer (2003) model also suggests that style momentum could be time-varying. I condition style momentum returns on January, lagged market state, lagged monetary policy changes and lagged changes in relative dispersion and find significant conditional style-level momentum. In the third essay, I identify and test explanatory factors that potentially predict style momentum returns. Several macroeconomic, relative dispersion, market related and volatility related factors are associated with future short-term style momentum returns. Interactions of many of these variables with market and monetary state indicator variables are significant in the regressions as well.

Fluid mixing that is induced by hydrodynamic instability is ubiquitous in nature; the material interface between two fluids when perturbed even slightly, changes shape under the influence of hydrodynamic forces, and an additional zone called the mixing layer where the two fluids mix, develops and grows in size. This dissertation reports a study on the role of thermal fluctuations in fluid mixing at the interface separating two perfectly miscible fluids of different densities. Mixing under the influence of two types of instabilities is studied; the Rayleigh-Taylor (RTI) and Richtmyer-Meshkov (RMI) instabilities. The study was conducted using numerical simulations after verification of the simulation methodology. Specifically, fluctuating hydrodynamic simulations were used; the fluctuating compressible Navier-Stokes equations were the physical model of the system, and they were solved using numerical methods that were developed and implemented in-house. Our results indicate that thermal fluctuations can trigger the onset of RTI at an initially unperturbed fluid-fluid interface, which subsequently leads to mixing of multi-mode character. In addition we find that for both RMI and RTI, whether or not thermal fluctuations quantitatively affect the mixing behavior, depends on the magnitude of the dimensionless Boltzmann number of the hydrodynamic system in question, and not solely on its size. When the Boltzmann number is much smaller than unity, the quantitative effect of thermal fluctuations on the mixing behavior is negligible. Under this circumstance, we show that mixing behavior is the average of the outcome from several stochastic instances, with the ensemble of stochastic instances providing the bounds on mixing-related metrics such as the mixing width. Most macroscopic hydrodynamic systems fall in this category. However, when the system is such that the Boltzmann number is of order unity, we show that thermal fluctuations can significantly affect the mixing behavior; the ensemble-averaged solution shows a departure from the deterministic solution. We conclude that for such systems, it is important to account for thermal fluctuations in order to correctly capture their physical behavior.

The northern hardwood forest type is a critical habitat component for the endangered Carolina northern flying squirrel (CNFS; Glaucomys sabrinus coloratus) for denning sites and corridor habitats between montane conifer patches where the squirrel forages. This study examined terrain data, and patterns of occurrence for the northern hardwood forest type in the recovery areas of CNFS in western North Carolina and southwestern Virginia with the purpose of creating a more robust predictive model of this forest type for spatial delineation. I recorded overstory species composition as well as terrain variables at 338 points throughout the study area in order to quantitatively define the northern hardwood forest type. These data were used in conjunction with digital terrain data for creation of the predictive model. Terrain variables we examined to attempt to differentiate northern hardwoods from other forest types included elevation, aspect, slope gradient, curvature, and landform index. I used an information-theoretic approach to assess six models based on existing literature and a global model.  My results indicate that on a regional, multi-state scale, latitude, elevation, aspect, and landform index (LFI) of an area are significant predictors of the presence of the northern hardwood forest type in the southern Appalachians.  My model consisting of Elevation + LFI was the best approximating model based on lowest AICc score.  Our Elevation + LFI model correctly predicted northern hardwood presence at 78.2% of our sample points observed to be northern hardwoods. I then used this model to create a predictive map of the distribution of the northern hardwood forest type in CNFS recovery areas. / Master of Science

Despite the emergence of artificial intelligence (AI) and machine learning (ML) in medicine and the resultant interest in predictive analytics in surgery, there remains a paucity of research on the actual impact of prediction models and their effect on surgeons’ risk assessment of post-surgical complications. This research evaluated how spinal surgeons predict post-surgical complications with and without additional information generated by a ML predictive model. The study was conducted in two stages. In the preliminary stage an ML prediction model for post-surgical complications in spine surgery was developed. In the second stage, a survey instrument was developed, using patient vignettes, to determine how providing ML model support affected surgeons’ predictions of post-surgical complications. Results show that support provided by a ML prediction model improved surgeons’ accuracy to correctly predict the presence or absence of a complication in patients undergoing spinal surgery from 49.1% to 54.8% (p=0.024). It is clear that predicting post-surgical complications in patients undergoing spinal surgery is difficult, for models and experienced surgeons, but it is not surprising that additional information provided by the ML model prediction was beneficial overall. This is the first study in the spine surgery literature that has evaluated the impact of a ML prediction model on surgeon prediction accuracy of post-surgical complications.

In this study, we use data mining techniques to build predictive models on data collected by a Portuguese bank through a term savings product campaign conducted between May 2008 and November 2010. This data is imbalanced, given an observed take-up rate of 11.27%. Ling et al. (1998) indicated that predictive models built on imbalanced data tend to yield low sensitivity and high specificity, an indication of low true positive and high true negative rates. Our study confirms this finding. We, therefore, use three sampling techniques, namely, under-sampling, oversampling and Synthetic Minority Over-sampling Technique, to balance the data, this results in three additional datasets to use for modelling. We build the following predictive models: random forest, multivariate adaptive regression splines, neural network and support vector machine on the datasets and we compare the models against each other for their ability to identify customers that are likely to take-up a term savings product. As part of the model building process, we investigate parameter permutations related to each modelling technique to tune the models, we find that this assists in building robust models. We assess our models for predictive performance through the use of the receiver operating characteristic curve, confusion matrix, GINI, kappa, sensitivity, specificity, and lift and gains charts. A multivariate adaptive regression splines model built on over-sampled data is found to be the best model for predicting term savings product takeup.

The purpose of this thesis was to compare two integration’s of sensors, into a system called Enlight, but could in theoryhave been integrated to most systems. As a pre-study, the specifications and availability of five sensors were researched.From the pre-study, Smart Edge 4.0 and Phantom EPH-V11/10 from Erbessd, were chosen and then integrated. Usability andperformance of the integrations were then compared usingcognitive dimensions and stopwatch. Phantom from Erbessdwas deemed to be more usable, and the integration of SmartEdge 4.0, had better performance.

Motivated by the goal of transition to a zero-carbon-emission-based economy for climate change mitigation, electrification opportunities are more promising in the transportation sector. Electric Vehicles (EVs) are at the forefront of the energy transition at an expanded rapid pace in the transportation sector. To enable and enhance the energy efficiency, advanced control and optimization will play an important role in EV systems and infrastructure. However, there are also some difficulties and limitations subject to the imperfection of management and control for EVs. Overall, to further the widespread adoption of EVs, the dissertation mainly includes two parts: 1) Power management for Plug-in Hybrid Electric Vehicles (PHEVs); 2) Charging control for Plug-in Electric Vehicles (PEVs). Chapter 2 deals with the power management and route planning problems for PHEVs, which aims to properly design the control algorithm to find the route that leads to the minimum energy consumption. Chapter 3 pays attention to the high workloads of the PEV in the electric power grids, which concentrates on studying a control algorithm leading to possible reductions in both computation and communication. Chapter 4 focuses on the charging control for PEVs, which explores how to improve the PEV charging efficiency while satisfying safety concerns. Chapter 5 modifies the results in Chapter 4 by taking battery capacity degradation into the optimization problem. This dissertation proceeds with Chapter 1 by reviewing the state-of-the-art control methods for PEVs and PHEVs. Chapter 2 studies a novel control scheme of route planning with power management for PHEVs. By considering the power management of PHEVs, we aim to find the route that leads to the minimum energy consumption. The scheme adopts a two-loop structure to achieve the control objective. Specifically, in the outer loop, the minimum energy consumption route is obtained by minimizing the difference between the value function of current round and the best value from all previous rounds. In the inner loop, the energy consumption index with respect to PHEV power management for each feasible route is trained with Reinforcement Learning (RL). Under the RL framework, a nonlinear approximator structure, which consists of an actor approximator and a critic approximator, is built to approximate control actions and energy consumption. In addition, the convergence of value function for PHEV power management in the inner loop and asymptotical stability of the closed-loop system are rigorously guaranteed. Chapter 3 investigates the self-triggered Model Predictive Control (MPC) with Integral Sliding Mode (ISM) method of a networked nonlinear continuous-time system subject to state and input constraints with additive disturbances and uncertainties. Compared with the standard MPC strategy, the proposed control scheme is designed for PEV charging to reduce the high communication loads caused by a large-scale population of vehicles under centralized charging control architecture. In the proposed scheme, the constrained optimization problem is solved aperiodically to generate control signals and the next execution time, leading to possible reductions in both computation and communication. The motivation of using ISM approach is to reject matched uncertainties. A self-triggered condition that involves a comparison between the cost function values with different execution periods is derived. Besides, the robust MPC with ISM control strategy is rigorously studied depending on the self-triggered scheme. Chapter 4 proposes a charging control algorithm for the valley-filling problem, while it meets individual charging requirements. We study a decentralized framework of PEV charging problem with a coordination task. An iterative learning-based model predictive charging control algorithm is developed to achieve the valley-filling performance. The design of the decentralized MPC meets individual charging requirements. The iterative learning method approximates the electricity price function and the system state sampled safe set to improve the accuracy of optimization problem calculations. The decentralized problem, in which the individual PEV minimizes its own charging cost, is formulated based on the sum of all power loads. Chapter 5 studies a modified charging control algorithm based on the previous charging control algorithm in Chapter 4. We propose a charging control algorithm for PEVs using a decentralized MPC framework supplemented by the iterative learning method. By considering the battery aging of PEVs, we aim to find the optimal charging rate that leads to valley-filling performance. The scheme adopts the iterative learning-based method to solve the optimal control problem with the battery aging model. Specifically, the sampled safe set and price function are updated accordingly as the iteration number increases. The battery aging model involves the cost function to approach the real charging scenario. In addition, the recursive feasibility of the proposed optimal control problem for PEV charging with battery aging and asymptotical stability of the closed-loop system are rigorously studied. Finally, in Chapter 6, the conclusions of the dissertation and some avenues for future potential research are presented. / Graduate / 2023-04-07

In recent years, the American Association of Colleges of Pharmacy (AACP) has encouraged the application of big data analytic techniques to pharmaceutical education. Indeed, the 2013-2014 Academic Affairs Committee Report included a "Learning Analytics in Pharmacy Education" section that reviewed the potential benefits of adopting big data techniques.1 Likewise, the 2014-2015 Argus Commission Report discussed uses for big data analytics in the classroom, practice, and admissions.2 While both of these reports were thorough, neither discussed specific analytic techniques. Consequently, this commentary will introduce classification trees, with a particular emphasis on their use in admission. With electronic applications, pharmacy schools and colleges now have access to detailed applicant records containing thousands of observations. With declining applications nationwide, admissions analytics may be more important than ever.3.

Modern synthetic gene regulatory networks emerge from iterative design-build-test cycles that encompass the decisions and actions necessary to design, build, and test target genetic systems. Historically, such cycles have been performed manually, with limited formal problem-definition and progress-tracking. In recent years, researchers have devoted substantial effort to define and automate many sub-problems of these cycles and create systems for data management and documentation that result in useful tools for solving portions of certain workflows. However, biologists generally must still manually transfer information between tools, a process that frequently results in information loss. Furthermore, since each tool applies to a different workflow, tools often will not fit together in a closed-loop and, typically, additional outstanding sub-problems still require manual solutions. This thesis describes an attempt to create a tool that harnesses many smaller tools to automate a fully closed-loop decision-making process to design, build, and test synthetic biology networks and use the outcomes to inform redesigns. This tool, called Phoenix, inputs a performance-constrained signal-temporal-logic (STL) equation and an abstract genetic-element structural description to specify a design and then returns iterative sets of building and testing instructions. The user executes the instructions and returns the data to Phoenix, which then processes it and uses it to parameterize models for simulation of the behavior of compositional designs. A model-checking algorithm then evaluates these simulations, and returns to the user a new set of instructions for building and testing the next set of constructs. In cases where experimental results disagree with simulations, Phoenix uses grammars to determine where likely points of design failure might have occurred and instructs the building and testing of an intermediate composition to test where failures occurred. A design tree represents the design hierarchy displayed in the user interface where progress can be tracked and electronic datasheets generated to review results. Users can validate the computations performed by Phoenix by using them to create sets of classic and novel temporal synthetic genetic regulatory functions in E. coli. / 2016-12-31T00:00:00Z

Der Einsatz modularer Anlagen wird in der Prozessindustrie immer beliebter. Da Modulhersteller ihre Applikationen weitestgehend kapseln, um ihr Know-How zu schützen, müssen Strukturen und Applikationen entwickelt werden, die das Monitoring und die Diagnose auch im Rahmen der vorausschauenden Instandhaltung dieser Module ermöglichen. Die Implementierung solcher Funktionen in das übergeordnete Leitsystem der Gesamtanlage ist bei modularen Anlagen bislang ähnlich umständlich wie die Einbindung von Fremdkomponenten in ein Leitsystem. Um ein vollständiges modulares Anlagenkonzept erfolgreich zu etablieren, ist eine einfache Implementierung dieser Funktionalitäten somit unumgänglich. Im Rahmen dieses Beitrags wird daher untersucht, wie Strukturen und Applikationen zum Monitoring und Diagnose einfach in übergeordnete Systeme implementiert werden können. Hier wird auch betrachtet, welche dieser Informationen für Anlagenfahrer relevant, welche Funktionalitäten auf externe Systeme (z. B. eine Cloud) ausgelagert werden sollten und wie diese optimal dargestellt werden. Weiterhin wird ein Konzept zur Modularisierung von Plant Asset Management Funktionen vorgestellt, anhand dessen eine Strukturierung des NAMUR Open Architecture (NOA)-Kanals vorgenommen wird.