Search

Noncompliance to treatment assignment is widespread in randomized trials and presents challenges in causal inference. In the presence of noncompliance, the most commonly estimated effect of treatment assignment, also known as the intent-to-treat (ITT) effect, is biased. Of interest in this setting is the complier average causal effect (CACE), the ITT effect among compliers. A further complication arises when the outcome variable is partially observed. My research focuses on estimating the distribution of a site-specific CACE in a multisite randomized controlled trial (MRCT) by maximum likelihood (ML). Assuming compliance missing at random (MAR). We express the likelihood as an integral with respect to random effects that cannot be evaluated analytically. We derive ML estimators by a combination of the EM algorithm and Newton Raphson method conditional on random effects and, then, integrate each estimator and the likelihood with respect to the random effects by adaptive Gauss Hermite quadrature (AGHQ). Next, we extend this approach to both outcome and imperfect compliance MAR. A distinctive feature of the approach is to estimate a site-specific CACE and its variance over sites efficiently. We applied our method to data sets from two MRCTs with imperfect compliance: the e-assist intervention trial to assess the effectiveness of the e-assist intervention in promoting colorectal cancer screening outcome fully observed; and the National Study of Learning Mindsets trial to evaluate the effectiveness of a growth mindset program in improving academic achievement outcomes MAR among ninth-graders.

This thesis describes the adoption of P450BM3 mutants as biocatalysts during target molecule synthesis in different stages and the efforts toward the synthesis of the azepane core type structure of natural product balanol and the enantiomer of natural product bulnesene. Chapter 1 provides an introduction to the application of P450BM3 mutants as oxidation catalysts for various structures including simple cyclic amines and sesquiterpenes. A review of the significance and various synthetic studies toward natural products balanol and bulnesene was provided. Chapter 2 describes a short demonstration of the feasibility of further functionalisation of P450BM3 hydroxylated cyclic amines by synthesising the azepane core of natural product balanol. Finally, Chapter 3 details a short enantiospecific route to ent-dihydrobulnesene and an investigation into the use of P450BM3 mutants for late-stage oxidation of sesquiterpenes, along with a comparison of P450BM3-mediated oxidation profiles of a pair of enantiomers.

One of the key planning trends in recent decades has been the movement towards the decentralisation of power, and the democratisation of the decision-making process. Throughout the world, there has been a tendency for spatial and territorial planning power to devolve towards lower-tier authorities (such as parishes, town councils and community groups) and this trend has been observed both in England and Taiwan. However, there remains some uncertainty about identifying the determinants regarding how these lower-tier authorities use decentralised powers. In England, some researchers have suggested economic and political factors as being among the most influential. Although there is evidence of increased involvement of lower-tier authorities in spatial decision-making in Taiwan, there has been little discussion of the implications of this for power change in the territorial planning process. One of the main aims of the current research is to provide an overview of the determinants of how lower-tier authorities use decentralised powers to make spatial and territorial (S&T) plans. The research uses mainly qualitative research methods to undertake eight in-depth case studies of local communities - four in England and four in Taiwan. In the case studies in England, the research finds that social deprivation may encourage local communities to become involved in creating a Neighbourhood Plan (NP), as local residents are motivated to improve their living environment, but this may not necessarily lead to planning implementation. In Taiwan, however, although deprived communities face a complex range of socioeconomic challenges, there is little evidence that this has motivated lower-tier authorities or community groups to play an active role in the local planning process. The impact of the government-led local planning system remains the key determinant in the S&T planning process. While the England government's move towards decentralising planning power has created some beneficial conditions for lower-tier involvement in the planning process, these benefits have not been replicated in Taiwan, where - by contrast - current approaches have limited impact to stimulate community to playing an active role in S&T planning at community level. The current research confirms the importance of government legislation as a platform for empowering lower-tier authorities, as evidenced by the case studies in England. The study also argues that the decentralised planning power is most strongly affected by, in England, lower-tier authorities' planning governance and, in Taiwan, by the particular circumstances facing local residents in some areas. In both cases, the fundamental determinant is strongly linked with the legal planning framework. The research recommends regular training support for local communities in the S&T planning process in both England and Taiwan. Specifically for England, the paper develops an assessment framework for examining lower-tier planning governance, while for Taiwan, the paper recommends strengthening lower-tier authorities' inputs in undertaking S&T planning at local level.

This thesis aims to evaluate underpinning mechanisms of oral processing of polysaccharide thickeners with a rod-like conformation. Xanthan gum (XG) and scleroglucan (SG) are identified as key research targets, with the view to identify the differences and similarities between XG and SG and to assess the potential of SG utilisation as XG analogue in specialised nutrition applications, such as in oropharyngeal dysphagia. Specifically, hydrodynamic properties of SG have been investigated, including hydrodynamic and thermodynamic non-ideality, as well as molecular weight distribution. Furthermore, rheological and lubrication properties have been studied to elucidate complex rheological behaviour of both polysaccharides, as well as to probe their interactions with human whole saliva and bovine submaxillary mucin, used as a saliva-mimicking model. The rheological behaviour was probed using a combination of the steady shear flow, small amplitude oscillatory shear, first normal stress difference and capillary break-up rheometry techniques, whilst lubrication and friction behaviour was examined using the soft-contact tribology method. Sedimentation velocity in analytical ultracentrifuge (AUC) showed SG preparation was unimodal when concentration is higher than 0.75 mg/ml. However, it was resolved into two main components at lower concentrations with partial reversibility between two components. Sedimentation equilibrium experiments revealed that SG exhibits partially reversible self-association behaviour. The molecular weight ranges from ~3 × 106 g/mol to ~5 × 106 g/mol after thermodynamic non-ideality correction. The results of SEC-MALS showed that the rod-shape structure of SG may transfer into a structure with more flexibility at higher masses. When XG solution or SG solution are mixed with mucin, sedimentation velocity experiment showed no apparent interactions at low concentrations. At higher concentrations, XG and SG solutions of the same concentration have comparable steady shear flow profile and linear viscoelasticity spectra. At the same time, they show marked differences in the elongational flow behaviour, normal stress difference and lubrication behaviour. When XG and SG are dissolved in viscosified matrix (40% sucrose solution), the differences were minimised, confirming that chain relaxation dynamics at longer characteristic time is similar between XG and SG. The influence of saliva (or mucin) on complex rheological behaviour and lubrication of XG and SG was found to be minimal, which is consistent with the results of the sedimentation analysis showing no effect in low concentration solutions. The examined sets of rheological and tribological behaviour show new opportunities for the use of SG as a XG-replacing thickener in medical nutrition applications.

This PhD thesis includes 4 chapters. The first and second chapters focus on testing the liquidity risk and liquidity risk premium in the Chinese stock market. The third and fourth chapters focus on testing the intrinsic value in the Chinese stock market. The main findings are: the liquidity risk premiums show significant negative from 2012 to 2016. And the portfolio liquidity risk has a structural change after the split-share structure reform. Furthermore, results indicate a negative liquidity risk premium in period 1 (2005-2007) for state-owned companies, Shenzhen Stock Exchange and Shanghai stock exchange in period 1. I do empirical analysis from the efficient market hypothesis and behaviour finance aspects. Moreover, in chapter 3 and 4, we find Chinese investors also focus on value and long-term investment. It shows that RIP and DCFP have better predictive power than other multiples. We found evidence of co-integration between price and accounting fundamentals (RI or DCF), and DCF or RI correct for the disparity between price and fundamentals (RI or DCF) in either whole CSM or real estate companies. The threshold regression results show that the predictive power of the RIP improved after the financial crisis, RIP shows lower predictive power after bull-bear market in 2015. The DCFP has better predictive power after the financial crisis, the predictive powers of the DCFP become weak after the bull and bear market in 2015. Namely, their predictive power has stage change.

Time series can be found in various domains like medicine, engineering, and finance. Generally speaking, a time series is a sequence of data that represents recorded values of a phenomenon over time. This thesis studies time series mining, including transformation and distance measure, anomaly or anomalies detection, clustering and remaining useful life estimation. In the course of the first mining task (transformation and distance measure), in order to increase the accuracy of distance measure between transformed series (symbolic series), we introduce a novel calculation of distance between symbols. By integrating this newly defined method to symbolic aggregate approximation and its extensions, the experimental results show this proposed method is promising. During the process of the second mining task (anomaly or anomalies detection), for the purpose of improving the accuracy of anomaly or anomalies detection, we propose a distance measure method and an anomalies detection calculation. These proposed methods, together with previous published anomaly detection methods, are applied to real ECG data selected from MIT-BIH database. The experimental results show that our proposed outperforms other methods. During the course of the third mining task (clustering), we present an automatic clustering method, called AT-means, which can automatically carry out clustering for a given time series dataset: from the calculation of global average time series to the setting of initial centres and the determination of the number of clusters. The performance of the proposed method was tested on 10 benchmark time series datasets obtained from UCR database. For comparison, the K-means method with three different conditions are also applied to the same datasets. The experimental results show the proposed method outperforms the compared K-means approaches. During the process of the fourth mining task (remaining useful life estimation), all the original data are transformed into low-dimensional space through principal components analysis. We then proposed a novel multidimensional time series distance measure method, called as multivariate time series warping distance (MTWD), for remaining useful life estimation. This whole process is tested on the CMAPSS (Commercial Modular Aero Propulsion System Simulation) datasets and the performance is compared with two existing methods. The experimental results show that the estimated remaining useful life (RUL) values are closer to real RUL values when compared with the comparison methods. Our work contributes to the time series mining by introducing novel approaches to distance measure, anomalies detection, clustering and RUL estimation. We furthermore apply our proposed methods and related methods to benchmark datasets. The experimental results show that our methods are better than previously published methods in terms of accuracy and efficiency.

Currently, building energy conservation has turned into one of the most attractive topic worldwide as buildings contribute the most energy consumption than any other consumers and they also produce a large amount of greenhouse gas emissions. In the United Kingdom, the residential building becomes the biggest energy consumer, with nearly 30% of the total energy consumption. Among them, existing traditional old style households consume much more energy compared to those new build domestic properties that adopted various energy saving methods. Therefore, it indicates a great opportunity to cut down the energy usage and reduce carbon dioxide emissions from buildings by retrofitting conventional houses. A detailed background study and literature review were reported involving current building energy challenges, domestic energy demand features, conventional and low energy domestic properties and the retrofitting of buildings. Two case studies were conducted to investigate the performance of a Conventional House and a Passive House, respectively. A set of data sensoring and recording instrument was installed in each house to record the properties' electricity and gas consumption, indoor temperature, relative humidity and CO2 concerntration for more than one year. Compuational models for both dwellings to simulate their energy consumption and indoor environmental condition were developed by DesignBuilder software. A simulation study by applying Passive House standard to retrofit the Conventional House was conducted after the model validation, which included adding insulation, utilising triple glazing and improving gas boiler's efficiency. An investigation to minimise the risk of summer overheating in the Passive House was carried out to further enhance its performance. The main findings of this research are: for the Passive House, the building performance was remarkable; for the Conventional House, retrofitting by using passive energy saving methods is essential to improve its building performance for reducing the energy demand and increasing the indoor temperature. Through the building fabrics improvement in the model simulation, the primary energy demand of the Conventional House was reduced significantly by 78% compared to its original status, and the indoor temperature was increased to comfortable level to satisfy the residents' requirement. Overall, the results from this research indicates the advantage of house retrofitting by using passive energy saving methods in achieving lower energy consumption and good thermal comfort. The fingdings of this study can be used as a reference for the future UK conventional household retrofitting.

In the present work, the relationship between the microstructure and the corrosion behaviour of Al-Cu-Mg and Al-Cu-Li alloys with various thermomechanical conditions has been studied. The microstructural characterization of the AA2024-T351 Al-Cu-Mg alloy revealed that the constituent intermetallic particles, assigned as S-phase (Al2CuMg), θ-phase (Al2Cu) and alpha-phase (Al-Cu-Fe-Mn-(Si)), are present individually or together as clusters. Further, precipitates along with Mg/Cu segregations were detected along the grain boundary network. It was also revealed that the cold working to obtain T351 temper resulted in the heterogeneous distribution of grain-stored energy. It was revealed that selective dissolution of Mg from the S-phase particle results in the copper-rich S-phase remnant, which contributes to the conversion of its electrochemical property. As a result, the micro-galvanic coupling between the S-phase remnant and the alloy matrix leads to the development of trenching at its adjacent. Trenching was also found in the periphery of the θ-phase particle and the alpha-phase particle, due to their more positive electrode potentials relative to that of the alloy matrix. Further, selective dissolution also occurred in the θ-phase and alpha-phase particles, resulting in the development of porous banding structure along certain orientation. Localized corrosion in the AA2024-T351 aluminium alloy preferentially propagated in the form of intergranular corrosion at the early stage of the exposure to sodium chloride solution. The distribution of grain-stored energy significantly affects the development of intergranular corrosion. With prolonged exposure time, localized corrosion propagated selectively into grain interior, resulting in the development of crystallographic pits. In the corrosion front area, the necessary chemical condition (low pH and chloride rich) was generated and maintained. Meanwhile, a copper-enriched layer beneath the corrosion product layer acted as an effective cathode to support the anodic dissolution. Therefore, the anodic dissolution of aluminium at the corrosion front is self-supported. The 2A97 Al-Cu-Li alloys in different thermomechanical conditions exhibited different amounts and distributions of T1 phase (Al2CuLi) precipitates since the pre-ageing cold working and the ageing condition significantly affect the T1 phase precipitation in Al-Cu-Li alloys. Further, the different thermomechanical histories also resulted in significantly different grain structures in the alloys. The corrosion morphology of the 2A97 alloy is closely associated with the distribution of T1 phase precipitates. In the T3 condition, attacked grain boundaries are the dominant corrosion features since T1 phase precipitates distribute along selective grain boundaries. In the T4 condition, localized corrosion propagated in the form of intergranular corrosion and developed into the grain interior with prolonged immersion time, related to the distribution of T1 phase precipitates. In contrast, crystallographic pits are evident in the grain interior of 2A97-T6 alloy due to the high population density of matrix T1 phase precipitates. Both grain interior and grain boundary were selectively attacked in the 2A97-T8 alloy during the immersion testing, which is consistent with the T1 phase precipitates distribution. Further, it was also found that the selective corrosion behaviour in the 2A97 Al-Cu-Li alloys is closely associated with the heterogeneous distribution of grain-stored energy, with high localized corrosion susceptibility corresponding to high level of stored energy.

In the current study, copper nanowires are fabricated by filling the nanochannels of anodic aluminium oxide (AAO) templates using electrodeposition. The nanowires with different growth orientations as well as twin boundaries in some cases are obtained. Cu micropillars are machined from bulk materials using focused ion beam (FIB) milling. The micropillars oriented in the [235] and [156] directions, with and without a twin boundary, with a cylindrical or square shape and with different sizes are prepared. These pillars are then compressed by a nanoindenter equipped with a flat tip, and the stress-strain curves are obtained. The morphologies of the pillars are observed using scanning electron microscopy (SEM). Thin foils of the pillars are prepared by FIB and examined using transmission electron microscopy (TEM) and scanning TEM (STEM). Finally, the dislocation structures are analysed to shed light on the investigation of the deformation mechanisms.

This work aims to develop infrared detectors and to introduce a new measurement technique for infrared radiation thermometry. It consists of two-colour detectors for ratio thermometry, InAs photodiode for 3.43 m narrow band thermometer and photon counting thermometer using a Si single photon avalanche photodiode (SPAD). In addition to research in these detectors, a Monte Carlo model for modelling impact ionisation in Si was also developed. InGaAlAs is attractive material for multi-colour detection at wavelengths up to 1.7 m, as it is lattice matched to InP substrate. InGaAlAs-InGaAs two-colour detector was evaluated as a ratio thermometer. When compared to a commercial Si-InGaAs detector, the InGaAlAs diode produces slightly higher (lower) output than Si at temperature below (above) 500 °C, while the InGaAs diode in this work also produces slightly higher output than that in the commercial Si-InGaAs detector. The InGaAlAs and InGaAs diodes detect blackbody temperatures as low as 275 and 125 oC, respectively, with signal to noise ratios (SNRs) above 10. As a ratio thermometer, the two-colour InGaAlAs-InGaAs photodetector achieves a temperature error of 12.8 °C at 275 °C, but this improves with temperature to 0.1 °C at 450 °C. If the maximum temperature error of 2 °C is defined, the InGaAlAs-InGaAs is capable of detecting an object temperature down to 325 °C. These results demonstrate the potential of InGaAlAs-InGaAs two-colour photodetector for development of high performance two-colour array detectors for radiation thermometry and thermal imaging of hot objects. The InAs photodiode offers huge potential for infrared sensing applications at wavelengths above 1.7 m. The performance of InAs photodiode was evaluated for use in radiation thermometry at wavelengths beyond InGaAs photodiode. For uncooled InAs, it successfully measured a blackbody temperature of 50 oC with an acceptable error of 0.17 oC. In order to evaluate its performance as a 3.43 m narrow band thermometer, measurements were repeated with a narrow band filter. InAs was demonstrated to have lower temperature error than a commercial PbSe detector. The temperature error was 1.88 oC for InAs at 50 oC compared to 3.78 oC for PbSe. This suggests that InAs is ideally X. ZHOU III suited for applications requiring 3.43 m operating wavelength. Further improvement was achieved by cooling InAs to 200 K. It was found that a temperature as low as 37 oC, with an error of less than 0.5 oC, can be measured indicating its potential for human body temperature sensing. An alternative to using a photodetector with longer wavelength response is to increase the sensitivity of the photodetector via internal gain mechanisms such as impact ionisation. By employing a very high internal gain in SPAD, the photon counting technique was evaluated for radiation thermometry. Photon induced avalanche pulses were successfully measured at temperature as low as 225 oC with an error less than 2 oC using Si SPAD. This is significantly lower than the lower temperature limit of 400 oC in conventional Si photodiode based radiation thermometer. The photon counting technique is therefore demonstrated to be a feasible technique to achieve lower temperature sensing.

This project was basically exploratory in the electrical properties of carbon nanotube (CNT) based materials. The direct current (DC) conductivity of CNT/polymer composites was computed by using equivalent circuit method and a three dimensional (3-D) numerical continuum model with the consideration of tunneling conduction. The effects of the potential barrier of polymer and the tortousity of CNTs on the conductivity were analyzed. It was found that both of percolation threshold and DC conductivity can be strongly affected by the potential barrier and the tortousity. The influence of contact resistance on DC conductivity was also computed, and the results revealed that contact resistance and tunneling resistance had significant influences on the conductivity, but did not affect the percolation threshold. The microstructure-dependent alternating current (AC) properties of CNT/polymer composites were investigated using the 3-D numerical continuum model. It was found that AC conductivity and critical frequency of CNT/polymer composites can be enhanced by increasing the curl ratio of CNTs. In the mid-range CNT mass fraction, with increasing curl ratio of CNTs, AC conductivity, interestingly, became frequency-dependent in low frequency range, which cannot be explained by reference to the percolation theory. A proper interpretation was given based on the linear circuit theory. It was also found that the critical frequency can also be affected by the size of CNT cluster. Series numerical formulas were derived by using a numerical capacitively and resistively junction model. In particular, this work introduced an equivalent resistor-capacitor (RC) circuit with simple definitions of the values of contact resistance and average mutual capacitance for CNT/polymer nanocomposites. Theoretical results were in good agreement with experimental data, and successfully predicted the effect of morphology on the AC properties of CNT/polymer composites. DC and AC conductivities of multi-walled carbon nanotube (MWCNT)/graphene oxide (GO) hybrid films were measured for selected MWCNT mass fractions of 10%, 33.3%, 50%, 66.7%, and 83.3% using four-probe method. The experimental results were fitted using scaling law, and relatively high percolation threshold was found. This high percolation threshold was understood in terms of the potential energy and intrinsic ripples and warping in the freestanding graphene sheets. The capacitance of these hybrid films were measured using the voltmeter-ammeter-wattmeter test circuit with different voltages and heat treatments. The MWCNT/GO film showed relatively high specific capacitance (0.192F/cm3 for the mass fraction of 83.3%) and power factor compared to conventional dielectric capacitors. Both of measured capacitance and power factor can be enhanced by increasing testing voltages. The capacitance of MWCNT/GO films rapidly decreased after heat treatments above 160 ℃. This decrease was caused by redox reaction in the GO sheets. The capacitive behaviour of MWCNT/GO hybrid films was also interpreted by using the equivalent circuit model. Single-walled carbon nanotube (SWCNT) and SWCNT/Poly(vinyl alcohol) (PVA) films were used to form a piezoresistive strain sensor. Both of static and dynamic strain sensing behaviours of SWCNT and SWCNT/PVA films were measured. It was found that the sensitivities of these films decreased with increasing their thicknesses. The SWCNT film with a thickness of 1900 nm and SWCNT/PVA film exhibited viscoelastic sensing behaviour, because van der Waals attraction force allowed axial slippages of the smooth surface of nanotubes. A numerical model was derived based on the dynamic strain sensing behaviour. This model could be useful for designing CNT strain sensors. Finally, thermoelectric power (TEP) of deformed SWCNT films with various thicknesses was measured. It was observed that positive TEP of SWCNT films increased with increasing stain above the critical point. The experimental results were fitted by using a numerical model in terms of a variation of Nordheim-Gorter relation and fluctuation induced tunneling (FIT) model. From the numerical model, it was found that the increase of TEP above the critical strain resulted from the positive term of the contribution from the barrier region, and the effect of barrier regions decreases with increasing the thickness of the film.

The drug discovery and development subsector lies at the heart of the pharmaceutical and biotechnology industry. However, previous studies have not distinguished this subsector from the industry as a whole. Little detailed analysis has looked at the fIrms that discover and develop new therapies. From a perspective of the Sectoral Systems of Innovation, this thesis aims to address this important gap in knowledge by looking at the structure, clustering, knowledge production and networking in the drug discovery and development subsector, and to stress the relevant policy implications. This study intends to objectively examine the best available indicators for the knowledge produced by this subsector and industry dynamics, therefore a broad design of methodology was chosen. Data was collected from government databases, scientific databases, commercial databases, industry associations and companies' websites, concerning the subsector's structure, clustering and concentration, research and development (R&D) investments, product pipelines, scientific publications and citations, patent publications, and alliance agreements. This study indicates that the drug discovery and development subsector was geographically clustered. The finding further reveals this subsector's hierarchical structure and divergence in strategy development. This thesis also suggests that the focuses of knowledge production in this subsector were changed when partners changed. Moreover, in arguing that this subsector featured massive knowledge production and expanding collaboration with other actors of the innovation systems, the analysis questioned the notion that domestic industry would benefit much from the successful knowledge production of this subsector, because much of the knowledge produced by this sector was going abroad through commercial licensing, and through mergers and acquisitions (M&As). The data of this study also indicated that the strategies of companies are co-evolved with its position within the networking and industry structure.

The aim of this thesis is to design and implement an advanced control system for a working three-phase DC to AC power converter. Compared to the traditional PI controller used widely in industry, the new voltage controller can track the reference voltage with improved accuracy and efficiency in the presence of different kind of local loads, and also works well in the single phase voltage control. This voltage controller is combined with a power controller to yield a complete controller. An important aspect of this work is the hardware implementation of the whole system. Main parts of this thesis are: 1. Review of H-infinity and repetitive control techniques and their applications in power converters. 2. Design of a new voltage controller to eliminate the DC component in the output voltages, and taking into account the practical issues such as the processing delay due to the digital signal processor (DSP) implementation. 3. Modelling and simulation of the converter system incorporating different control techniques and with different kinds of loads. 4. Hardware implementation and the two-processor controller. The parallel communication between the DSPs. 5. The main problems encountered in hardware implementation and programming. The software used to initialize DSPs, implement the discrete time voltage controller and other functions such as generations of space vector pulse width modulation (SVPWM) signals, circuit protections, analog to digital (AD) cOl)versions, data transmission, etc. 6. Experimental results under circumstances of no load connected to the converter, pure three-phase resistive loads, three-phase unbalanced resistive loads and the series resistor-inductor loads.

This paper presents Chinese art songs from the early 20th century to today and highlights selected composers from each of the five traditional eras of recent Chinese social history. It is intended as an introductory guide for singers and collaborative pianists who are interested in Chinese art songs and the development of the genre in China. Selected composers include Youmei Xiao (1884-1940), Yuanren Zhao (1892-1982), Er Nie (1912-1935), Hanhui Zhang (1902-1946), Xinghai Xian (1905-1945), Deyi Shang (1932-2020), Zhongrong Luo (1924-2021), Zaiyi Lu (b. 1943), Jianfen Gu (b.1935), and Cong Liu (b.1956). The paper provides brief biographical information for each composer and lists their selected art songs and a discography of available recordings.

New computing applications, e.g., deep neural network (DNN) training and inference, have been a driving force that changed the semiconductor industry landscape. The data-intensive nature of DNN applications usually leads to high computation costs and complexity, and different hardware accelerators have thus been developed to improve the efficiency of running these models. For example, NVIDIA GPU, google TPU and near-memory computing architectures can enhance DNN performance and energy efficiency compared to conventional CPUs. Past that, IMC methods can circumvent the fundamental von Neumann bottleneck and enable highly parallel computing, leading to even higher hardware efficiency and performance. This thesis examines several aspects of IMC accelerators based on emerging memory devices such as RRAM, which can offer high computation density, throughput and energy efficiency. We present a reconfigurable IMC design that can accelerate general arithmetic and logic functions. The system consists of small look-up tables (LUTs), a memory block, and search auxiliary blocks integrated into an RRAM crossbar array. External data access and data conversions are eliminated to allow operations fully in-memory. Details of logic and arithmetic functions such as addition, AND and multiplication are discussed based on search and writeback steps. A compact instruction set is demonstrated in this architecture through circuit-level simulations. Performance evaluations show that the proposed IMC architecture suits data-intensive tasks with low power consumption. Next, we discuss DNN accelerator designs using a tiled IMC architecture. Popular models including VGG-16 and MobileNet are successfully mapped and tested on the RRAM-based tiled IMC architecture. Effects of finite RRAM array size and quantized partial sums (Psums) due to ADC precision constraints are analyzed. Methods are developed to address these challenges and preserve DNN accuracy and IMC performance gains. For practical IMC implementations and to support larger models, we develop a Tiled Architecture for In-memory Computing and Heterogeneous Integration (TAICHI), a general IMC DNN accelerator design. TAICHI is based on tiled RRAM crossbar arrays heterogeneously integrated with local arithmetic units and global co-processors to allow the same chip to efficiently map different models while maintaining high energy efficiency and throughput. A hierarchical mesh network-on-chip is implemented to facilitate communication among clusters in TAICHI to balance reconfigurability and efficiency. Detailed implementation of the different circuit components is presented, and the system performance is benchmarked at several technology nodes. The heterogeneous design also allows the system to accommodate models larger than the on-chip storage capability to make the hardware system future-proof. Large-scale implementations of IMC accelerators face two technological challenges – high ADC overhead and device variability. We note these challenges can be addressed by restricting neuron activations to single-bit values, i.e., spikes, and by employing binary weights. Based on these principles, we propose efficient hardware implementation of binary-weight SNNs (BSNNs) that can be achieved using current RRAM devices and simple circuits. Binary activations also provide opportunities for intra and inter-layer data routing, and neuron circuit design optimizations. Through high-precision backpropagation-through-time (HP-BPTT) and a proper neuron design, we show BSNN can achieve accuracies comparable to floating-point models. With these co-designs, the proposed architecture can achieve high energy efficiency and accuracy for common SNN datasets. The robustness of the BSNN model against device non-idealities is further verified through experimental chip measurements. Finally, we discuss other opportunities to further enhance IMC architecture performance, including possible pipelining optimization, mapping strategy and BSNN training optimization strategies.

Photocatalytic water splitting to produce H2 fuel has drawn considerable attention due to ever-increasing environmental concerns and the rising global energy demand for renewable and clean energy formation. Since last century, the semiconductor photocatalysts have been studying to explore their photocatalytic properties for solar-driven H2 evolution from water. Among the various H2-evolving photocatalysts, CdS with excellent visible-light absorption and suitable band potentials is one of the most studied photocatalysts. Although cocatalyst-decorated CdS nanorods (NRs) offer a promising H2 evolution performance, further benefits invoked by spatial cocatalyst decoration are worth exploring. Given that noble metal cocatalysts were most deposited on the tips of CdS NRs, the tip deposition of non-noble metal cocatalysts on CdS NRs was explored in the first stage of PhD research. For the first time, it was found that amorphous MoOxSy preferentially photodeposited on CdS NR tips, which enhanced photocatalytic H2 evolution. Detailed characterization and studies determined the composition and possible formation pathways of the amorphous MoOxSy on the CdS tips. The enhancement was ascribed to the effective trapping of photoinduced electrons by the MoOxSy resulting from its lower surface work function compared to CdS. However, amorphous MoOxSy is unstable in most hole scavengers except methanol. Consequently, the MoOxSy was vulcanized into MoS2, with MoS2 presenting either on the only tips, on the tips and random walls, or having an overall coating. MoS2-tipped CdS NRs exhibited a better H2 evolution activity compared to other MoS2-loaded CdS NRs. Kelvin probe force microscopy and time-resolved photoluminescence spectroscopy were used to study spatial charge separation, transfer, and lifetime. The superior photoactivity of MoS2-tipped CdS NRs was assigned to spatially intensified charge separation along the long axis of NRs, thereby lowering the charge recombination rate. To further understand the effects of MoS2 distribution on CdS NRs, MoS2 was parallelly or perpendicularly placed on polar CdS (001) surfaces (representing CdS NR tips) and non-polar CdS (110) surface (representing CdS NR wall) to construct different CdS/MoS2 models for DFT calculations. It was verified that CdS NR is capable of the longitudinal charge transfer due to charge polarization on polar (001) surfaces. The distinct electronic properties (i.e. shift in core-level energies of Cd and S orbitals, interfacial charge accumulation and depletion) of different CdS/MoS2 heterostructures, induced by the different interfacial bonding arrangements and MoS2 orientation, were observed and analyzed. In addition, two types of MoS2-loaded CdS NRs (MoS2 being only on the tips or only on the walls of CdS NRs) were fabricated to obtain their BEs of Cd 3d and S 2p by XPS for comparison, which is consistent with the expected DFT results. The band edges of CdS component in the as-prepared samples were also determined by combining Tauc plots and VB XPS measurements; the band shifts in experiments approximately correspond to DFT results. As a result, the MoS2 distribution and orientation affect the electronic properties of CdS/MoS2 heterostructures and thereby giving a different photoactivity. Overall, this thesis reveals that the origins of photocatalytic H2 evolution advantages induced by spatially located cocatalyst and gives insights into the activity enhancement mechanism for tip-decorated CdS NRs, which guides the construction of more efficient CdS-based photocatalysts.

Abstract: Computerized formative assessments (CFA) are becoming popular in post-secondary education because they can offer immediate feedback and on-demand testing. This has created a pressing need for large numbers of content-specified items together with their feedback so that the items may be available continuously for formative assessment. To support CFA in post-secondary education, I present a modified generation framework to address the operational difficulties that arise when applying existing frameworks. I also evaluate the application of the modified framework for item generation within the context of higher education. The proposed framework incorporates a human-machine interactive approach and employs a tree structure for cognitive modeling as well as introduces a new mechanism to assemble elements and a validation tool. The new framework was applied within the university teacher education field. Specifically, two instructors of EDPY 303 (Educational Assessment) used HIM_AIG, which is a software program developed based on the modified framework, to implement generation tasks for quizzes used in the course. Six experts in higher education evaluated the generated content using the statements that have achieved the greatest consensus among published item-writing and feedback-writing guidelines/recommendations. One hundred and thirty-four students took the practice quizzes, which contained the generated items. The results suggest that this modified approach is feasible; the evaluative quality of the generated items and feedback is comparable with that of the parent ones; the psychometric quality of the generated items satisfies the standard. Implications of the study and limitations of this research are also presented.

Surface reconstruction and 3D modeling is a challenging task, which has been explored for decades by the computer vision, computer graphics, and machine learning communities. It is fundamental to many applications such as robot navigation, animation and scene understanding, industrial control and medical diagnosis. In this dissertation, I take advantage of the consumer depth sensors for surface reconstruction. Considering its limited performance on capturing detailed surface geometry, a depth enhancement approach is proposed in the first place to recovery small and rich geometric details with captured depth and color sequence. In addition to enhancing its spatial resolution, I present a hybrid camera to improve the temporal resolution of consumer depth sensor and propose an optimization framework to capture high speed motion and generate high speed depth streams. Given the partial scans from the depth sensor, we also develop a novel fusion approach to build up complete and watertight human models with a template guided registration method. Finally, the problem of surface reconstruction for non-Lambertian objects, on which the current depth sensor fails, is addressed by exploiting multi-view images captured with a hand-held color camera and we propose a visual hull based approach to recovery the 3D model.

The objective of this thesis is to develop a new method, re-parameterizing the logistic dose-toxicity model of the time-to-event continual reassessment method (TITE-CRM) for phase 1 dose-escalation cancer trials, to account for patient heterogeneity. In a phase 1 dose-escalation clinical trial, the goal is to identify a dose (the maximum tolerated dose or MTD) that is associated with a pre-specified probability of unacceptable toxicity. TITE-CRM is a Bayesian, model-based dose-escalation trial design specificially intended for situations where participants must be observed for toxicity a long period time compared to the rate at which patients present for treatment. The risk of toxicity may vary due to particpants’ inherent characteristics. For example, patients with lung cancer and concomitant renal disease are not able to tolerate the same dose level of the cancer treatment as the patients without concomitant renal disease. If patient heterogeneity exists, the one-group study design will ignore the patient heterogeneity and give an average MTD, which could result in excess toxicity in the higher risk patients and suboptimal dosing in the lower risk patients. We have developed a new method to address this issue by re-parameterizing the logisitic model of the TITE-CRM and adding dose-escalation rules to reflect risk group ordering information and control aggressive dose escalation. We assessed the operating characteristics of this design in simulations assuming three risk groups, although the model is trivially extensible to a greater number. We compared this method to parallel trials that use independent one-parameter logistic models. We investigated scenarios with equal numbers of patients in the risk groups and more patients in the higher toxicity risk group, three different true dose-toxicity models and three sample sizes, for a total of 5,400 trials. According to the results, the TITE-CRM with re-parameterized logistic model used in 3-group trials performs similarly to the TITE-CRM with one-parameter logistic model used in the parallel trials in terms of operating characteristics. However, the TITE-CRM with the re-parameterized logistic model worked better in terms of in-trial dose allocation of patients and recommending the correct final dose in the scenario where there were more patients with higher toxicity risk in a trial. However, estimations of the group risk difference were not precise and the coverages of credible intervals of the group difference parameters were excessively conservative. The method shows promise, and could be implemented now, but further improvements are required. Cancer is one of the major public health concers in the modern society and the new method can be used in the phase 1 dose-escalation cancer trial and have positive impact on the public health.

This manuscript explores the relationships among identity, media, and popular music in Chinesesociety through an examination of a televised singing competition franchise, The Voice of China.I attempt to understand what role popular culture, in the form of a contemporary popular culturalproduct, plays in Chinese people’s everyday life and how the show is a site where Chinesepeople articulate, interrogate and negotiate aspects of identity. I provide a textual analysis tointerpret the meaningful details of each episode while employing a critical-cultural approach tounderstand the socio-historical contexts and the online discourse that are conducive to theuniqueness of the international franchise. More specifically, I examine how societal discourseson the blind audition and the voting systems of the show reveal growing public awareness of andconcern with issues of equity and fairness in the cultural arena of traditional aesthetic standardsas well as the societal arena of electoral politics. I also explore how the show reinforces theruling Party’s appropriations of nationalism in a sophisticated way to highlight a unified andsupreme national voice, how it reflects both a loosening grip of the central government on genderrepresentation and a growing social leniency toward gender diversity, and finally how the voicediversity was constructed by individuals who are in possession of or in need of various forms ofsocial capital and who are from geographically diverse and economically disparate socialbackgrounds. I point out that The Voice of China offers a prime site to understand how thediscourses of politics, (trans)nationality, ethnicity, gender, and class intersect with each other inthe situated context and hopefully, such a discussion can raise the cultural awareness tochallenge institutional power and social norms in order to hear the diversified voices in Chinesesociety.

Diva presents a look at women in the present day. The inspiration for this thesis project comes from rhythmic gymnastics, a unique sport exclusively for women. The movement of the ribbon, twining and surrounding a woman’s body, highlights the beauty of the woman. The twining movement symbolizes the complex and developing roles women are playing in contemporary society. The structures of the pieces in this thesis project are open and vast, creating a sense of allure, as well as a reflection of the open mind of the contemporary woman. The pieces are separated into two parts, black and white, symbolizing different characteristics of women. Black and white are classic and basic colors, but opposites. Neon pink, a modern pop color associated with women, relates to femininity and links the black and white together, symbolizing different kinds of women in the modern world.

What kind of market structure promotes innovation and growth? This dissertation delves into the relationship between market power, innovation, and returns. On one hand, competition exerts downward pressure on costs and provides incentives for efficient organization of production. On the other, the Schumpeterian hypothesis suggests that monopolist rents are necessary to encourage technological disruption. I study the effects of competition in two distinct settings: the acquisition of early-staged high-growth startups and horizontal mergers between public companies. In Chapter 1, Catering Innovation: Entrepreneurship and the Acquisition Market, I study the role of the financial market of acquisitions on an inventor's decision to begin a new venture and his or her subsequent innovation. After documenting its increasing importance as the dominant exit path for entrepreneurs, I test a novel catering theory of innovation: Does the market structure of potential acquirers have a measurable impact on inventors' start-up decisions? I construct a new dataset of early stage start-ups using the uniquely broad coverage of CrunchBase and VentureXpert data. I disambiguate and match the resulting data to employment data from LinkedIn and to the entire universe of patent data. Using the prior citation history of entrepreneurs for exogenous variation, I construct a formal proxy variable and employ the Heckman selection model to establish causality. I find that a one standard deviation increase in acquirer market concentration decreases the inventor's propensity to become an entrepreneur by 4%. This first result suggests that fragmented markets are appealing entry markets. My main finding is that a one standard deviation increase in acquirer concentration and market size increases the quality of patents, as measured by citations per patent, and the catering of entrepreneurs, as measured by technological overlap with potential acquirers. The magnitudes suggest that 5-16% of entrepreneurial innovation can be attributed to the influence of acquisition markets, particularly in the information technology and biotechnology industries. In Chapter 2, Market Power in Mergers and Acquisitions, I explore the value implications of market power changes in public mergers and acquisitions. Using a large sample of horizontal mergers from 1980 to 2012, I provide evidence that a significant portion of merger announcement returns are explained by changes in market power resulting from the merger. I find that a 0.1 expected increase in product market concentration leads to a 2.3% increase in cumulative abnormal returns over a three day period. In the long-run, mergers with high expected changes in market power revert to pre-announcement levels. My results suggest the "announcement effect" in M&A is due to misplaced market power expectations by investors.

Most of today’s mobile apps rely on the underlying networks to deliver key func- tions such as web browsing, file synchronization, and social networking. Compared to desktop-based networks, mobile networks are much more dynamic with frequent con- nectivity disruptions, network type switches, and quality changes, posing unique pro- gramming challenges for mobile app developers.As revealed in this thesis, many mobile app developers fail to handle these inter- mittent network conditions in the mobile network programming. Consequently, network programming defects (NPDs) are pervasive in mobile apps, causing bad user experiences such as crashes, data loss, etc. Despite the development of network libraries in the hope of lifting the developers’ burden by providing flexible fault-tolerant APIs, we observe that many app developers often ignore such APIs or lack the capability to use them correctly and therefore still introduce NPDs.In this thesis we make three contributions towards robust network programming for mobile apps. In the beginning, we study the characteristics of 90 real-world NPDs in Android apps towards a deep understanding of their impacts, root causes, and code patterns. This study exposes the common mistakes and pitfalls made by app developers when dealing with intermittent network. It also reflects that NPDs are pervasive despite the availability of fault-tolerant API from third-party network libraries.Driven by the study, we build NChecker, a practical tool to detect NPDs by stat- ically analyzing Android app binaries. NChecker has been applied to hundreds of real Android apps and detected 4180 NPDs from 285 randomly-selected apps with a 94+% accuracy. Shockingly, NChecker detects network bugs in 99% of the evaluated apps. Our further analysis of these defects reveals the common mistakes of app developers in working with the existing network libraries’ abstractions, which provide insights for improving the usability of mobile network libraries.Besides NPD detection, a more fundamental way to eliminate NPDs is to pro- vide developers easy-to-use interfaces for network programmings. We argue that the difficulty of avoiding NPDs can be mitigated through an annotation language that al- lows developers to declaratively state desired and actual properties of the application, largely without reference to fault-tolerant concepts, much less implementation. A pre- compiler can process these annotations, replacing calls to standard networking libraries with customized calls to a specialized library that enhances the reliability. The last com- ponent of the thesis presents ANEL, a declarative language and middleware for Android that enables non-experts. We demonstrate the expressiveness and practicability of ANELannotations through cases studies on real-world networked mobile apps. We also show that the ANEL middleware introduces negligible runtime performance overhead.

As the second most leading cause of cancer death in women, breast cancer has attracted wide attention in MRI studies. However, the segmentation of non-mass lesions in breast cancer MRI images remains challenging due to their comparatively high variations in kinetic characteristics and typical morphological parameters. Meanwhile, the segmentation of spinal tumors in axial and sagittal views has been tough task because of the complexity to determine lesion area in one single view.In this thesis, we proposed a method for automatic non-mass tumor segmentation, which was threshold oriented and based on region growing algorithm. It turned out that the segmentation results were satisfactory where only less than 5% pixels compared to the total number of segmented tumor pixels were corrected. We also introduced a method for spinal tumor segmentation to determine the location of lesions in spinal cord cancers. Normalized cut and region growing were applied in this study so to locate spinal tumors. Seven types of spinal cancers were analyzed and the overall behavior of the method was good in most cases except for some cases due to influence of unclear boundaries of the lesion areas. Finally, we applied a machine learning based method to differentiate three subtypes of breast cancer tumors and three types of spinal cancers. ANOVA and random forest algorithm was utilized to select the most significant features and cross-validation-based logistic regression was then used in this step in order to build a classifier. The performance of the classifier can be improved if more cases are included and affecting factors like biomarkers in spinal cancers can be rejected from the classification process.

Abstract: The purpose of this study is to describe a methodology to recover the model (cognitive and item models) from generated test items using a novel graph theory approach. Beginning with the generated test items and working backward to recover the original model (cognitive and item models) using a systematic process with graph theory serves as model-based method for validating automatically generated test items. The methodology is demonstrated using generated items from the medical education domain. The proposed methodology was found to be systematic and generalizable using three different datasets.

New hetero-structured La2NiO4+delta (LNO)/ (La0.6Sr0.4)0.95Co0.2Fe0.8O3-delta (LSCF) SOFC cathodes have been successfully fabricated by the infiltration technique for long-term stability and oxygen reduction reaction (ORR) kinetic consideration. The current comparative study of LSCF and LNO-infiltrated LSCF shows a significant reduction in polarization resistance by one order of magnitude and a 67% increase in maximum power density by using LNO-infiltrated LSCF cathodes at 750°C in air. The electronic conductivity relaxation (ECR) results demonstrate that the remarkable improved oxygen exchange properties of LNO/LSCF hetero-structured cathodes are attributed to not only the increase of surface reaction area, but also the introduction of LNO surface and LNO-LSCF interface with rapid oxygen exchange behavior. A trace amount of Co, Sr and more oxygen defects stabilized in the LNO layer have been identified by TEM, EDXS, XPS, XRD and iodine titration experiments, giving rise to the enhancement of stability and surface catalytic activity. A low degradation rate of 0.39% at a constant current density of 250 mAdeltacm-2 can be still achieved for the fuel cell using LNO-infiltrated LSCF cathodes after long-term durability of about 500 h at 750°C, accompanied by nanoparticles growth/aggregation, delamination and slight lanthanum enrichment on LNO surfaces. The benefits from the presence of the LNO nanoparticles demonstrate that LNO-infiltrated LSCF materials can act as high active surface oxygen exchange cathodes with promising fast ORR behavior and stability.;To further enhance the long-term stability of infiltrated (La0.6Sr0.4)0.95Co0.2Fe0.8O3-delta (LSCF) cathode, two kinds of new hetero-structured cathodes, CeO2 & LNO co-infiltrated LSCF and La2-xNiO4+delta-infiltrated LSCF, have been designed. La2NiO4+delta (LNO), as an superior performance promoter, and CeO2, as a particle growth retardant due to the high melting point and good wettability, were co-infiltrated into the LSCF backbone to mitigate the delamination and growth/aggregation of in-situ formed nano-sized infiltrants. A more advantageous microstructure was confirmed in the co-infiltrated sample relative to those infiltrated with LNO alone. The original morphology was largely retained in the co-infiltrated sample after heat treatment, showing finer surface particles, better connection between the infiltrant network and the LSCF backbone. XPS and EIS results reveal that the substitution of stoichiometric La2NiO4+delta by A-site deficient La2-xNiO4+delta can effectively suppress lanthanum enrichment and enhance the stability after cathodic polarization treatment observed by measurement. The advantages of both modified hetero-structured cathodes suggest that CeO2 & La2-xNiO4+delta co-infiltrated LSCF is a promising hetero-structured cathode to mitigate the LSCF electrode degradation.

Large cities often become regional cities. This has become a global trend emerging in early 20th century. As the urbanised area grows, the connection with nature tends to weaken – therefore green space networks can play an important role in shaping regional urban form. New Zealand’s largest urban region, Auckland, is facing the challenges of increasing population and ongoing urban growth. This research project aims to preserve Auckland’s green network to retain Auckland’s unique lifestyle and ensure that the future growth is environmentally sustainable. By identifying the distinguishing characteristics of Auckland’s lifestyle and urban patterns, this research project takes the view that Auckland regional development should focus on the strong interrelationship between city and nature. Based on knowledge of other cities’ experiences, this research work not only includes design theories and techniques, but also incorporates the implementation policies necessary to achieve the goal of preserving Auckland’s green network. Research methods include defining appropriate environmental criteria and social criteria. Regional Auckland, the Warkworth-Silverdale (W/S) greenbelt and the Puhoi village were used as case studies to test this approach. Firstly, a set of environmental criteria was established in order to analyse green space features and identify potential green belts in regional Auckland. By combining all the important environmental conditions, six possible green belts were identified as buffers between urban developments. Secondly, the W/S greenbelt was used to show how the environmental criteria could be used at the local scale. After identifying the green network around the W/S greenbelt, social criteria (which were derived from an independent Remuera study) were used to shape Puhoi’s future urban structure inside the W/S greenbelt. Finally, a master plan was carried out to show how both sets of criteria, environmental and social, could be integrated to maintain a high quality lifestyle with great access to nature. The research findings suggest that a green network can, in addition to offering the growing population a new regional park system, also provides more urban land for the growing Auckland and ensures a high quality lifestyle for its future citizens.

Availability of labile nitrogen (N) is important in shaping composition, diversity, and dynamics of organisms and may impact ecosystem functioning in aquatic environments. Our knowledge on biogeochemical cycles of N and microorganisms that mediate these processes has been significantly improved in the past few decades. New findings, such as the discovery of anaerobic ammonium oxidation (anammox), an alternative route to remove bioavailable N, have shifted the traditional paradigm of nitrogen cycling. A new question, therefore, is raised on the contribution of anammox, relative to denitrification, in N2 production. Moreover, some long-standing questions, such as the chemical composition of labile dissolved organic nitrogen (DON) pool and the transformation mechanism of labile DON are yet to be solved. The overall objective of this research was to improve our understanding of bacterially mediated N transformations in aquatic environments, specifically on nitrogen removal by anammox and denitrification and on polyamine (a labile DON) transformation.

Joint modeling techniques have been developed for analyzing correlated longitudinal and survival data in many studies. It provides consistent and efficient estimates of the parameters even when the longitudinal covariate is measured infrequently and with measurement error. This work focuses on the use of joint modeling to solve two different statistical problems. The first one is about analyzing censored biomarker measurements and survival data under a case-cohort design. The goal is to study how the biomarker level changes over time, and the relationship between longitudinal biomarker measurements and the event time. We suggest a modified likelihood-based approach to adjust the possible bias introduced by the censoring due to detection limits and the measurement error in biomarker measurements. The second topic is about drawing inference for mean quality adjusted lifetime data. We consider continuous health experience and define the quality function with repeatedly measured quality of life score. We propose a consistent and asymptotic normal estimate for the mean quality adjusted lifetime and derive its asymptotic variance. The performances of the proposed methods have been demonstrated in simulation studies and through real data examples. Public Health Significance: A case-cohort study is a cost-effective design that is used in many large epidemiological studies. The method proposed in the first part of this dissertation increases the efficiency of the parameter estimation under case-cohort designs, which can lead to a considerable reduction in cost and effort. The evaluation of health benefits is a major public health challenge. The second part of this dissertation presents the estimate of mean quality adjusted lifetime, which can serve as a measurement of health benefits in the comparison of treatments or public health strategies.

Based on their different reproductive modes and evolutionary history, extant mammals are divided into three lineages: eutherians, marsupials and monotremes. Marsupials are the closest relatives to eutherians. They are born immune-incompetent. The characteristic that they develop their immune system postnatally makes them unique models to study newborn immune development and maternal immunology. As monotremes comprises the basal lineage of mammals, the study of monotreme immunology will provide unique insights regarding the origin of the mammalian immune system. Unfortunately, compared to our in-depth knowledge of eutherian immunology, especially humans and mice, knowledge of marsupial and monotreme immunology is scarce. The goal of the first part of this research was to investigate the development of immune-competence in a model marsupial, the gray short-tailed opossum, Monodelphis domestica. This research specifically looked at the ontogeny of B cells with emphasis on diversity of immunoglobulin genes during postnatal development. It takes advantage of genomic resources available for the opossum and a captive colony at the UNM Biology Department. To achieve these goals, the content and genomic organization of Ig heavy chain and light chain loci was determined. Opossum Ig heavy chain are of low germ-line diversity while light chains have high germ-line diversity. This suggests that opossums rely more on light chains than heavy chains for repertoire diversity (Wang et al. 2009). Using the detailed genomic information available of Ig loci, the timing of B cell ontogeny and Ig repertoire diversity was then determined. Opossum newborns start heavy chain VDJ recombination within the first 24 hours postpartum. The expression of the surrogate L chains occurs at day 6 postnatally. The subsequent rearrangement of the Ig\u03bb and Ig\u03ba L chain genes occur at days 7 and 8 postnatal, respectively. The diversity of early B cell H chains is limited and reduced in N region additions, as has been seen in fetal humans and mice, but lacks bias in the V, D and J segments used. Different from H chains, L chains develop much more diverse VJ recombinations and high IgL repertoire diversity when first expressed. Collectively the results demonstrate that B cell development is entirely postnatal in the opossum. The earliest time-point that an opossum has mature B cells is at the starting the second week of life (Wang et al. in preparation). These results are consistent with earlier work demonstrating that most marsupial species, including opossums are unable to generate an antibody response until the second week. A second goal of my research was the characterization of a novel T cell receptor in the duckbill platypus Ornithorhynchus anatinus. Although different from the previous goal, it nonetheless uses a non-eutherian model to address broader questions regarding immunoglobulin and T cell receptor gene evolution. TCRμ is a new T cell receptor that was first identified in marsupials and does not exist in eutherians (Parra et al. 2007). Homology searches of the platypus genome with opossum TCRμ sequence have identified a homologue of this unconventional TCRμ in platypus. Platypus TCRμ is expressed in a double V domains structure and these resemble Ig V more than conventional TCR V domains. Different from opossum TCRμ, platypus TCRμ requires two rounds of somatic recombination to assemble both V domains. The identification of TCRμ in platypus indicates that TCRμ as an ancient T cell receptor has been lost in eutherians.

The quantum spin Hall (QSH) phase exhibits spontaneous quantized (charge/spin) conductance in the absence of external magnetic field, as a result of strong spin-orbit coupling. However, due to the large bulk contribution in 3D TIs, it is difficult to observe these new states. Here we first demonstrate a selective control of two surfaces conductions (top/bottom) in epitaxially grown Bi2Te3 thin films by gate-dependent Shubnikov-de Hass oscillations. By sweeping the gate voltage, the Fermi level of the top surface is selectively tuned while the bottom surface conduction remains unchanged due to strong electric-field screening effects arising from the high dielectric constant of Bi2Te3. In addition, the total bulk conduction can be modulated from n- to p-type with a varying gate bias. We further investigate BixSb2-xTe3 as it provides lower bulk contribution and enhanced two surface conductions. These Sb-doped materials provide us a good opportunity to see the quantum spin Hall result.

Endometrial cancer is the third most common cause of gynecologic cancer death and shows the largest overall survival difference (34%) between the races. The National Cancer Institute (NCI) Black/White Cancer Survival Study was a population-based study of racial differences in cancer survival. Endometrial cancer cases consisted of 149 black women, ages 20-79 years, residing in three selected metropolitan areas, who were diagnosed with endometrial cancer between 1985 and 1987. Cases were frequency matched in a ratio of approximately 1:2 to a sample of 341 white women with endometrial cancer. Information was derived from abstracts of hospital and physicians' records, centralized pathology review, and interviews. Potential explanatory factors for black-white survival differences have been previously investigated using Cox regression. However, there was a high proportion of missing values since 24 percent of patients were never interviewed. Some values were also missing for three other variables derived from medical records. Missing values may introduced bias in previous findings based only on the information available.The primary objective of this thesis is to evaluate the effect of missing data on the estimated black/white mortality ratios adjusted for various explanatory factors. A second objective is to obtain more precise confidence intervals for the estimated mortality ratios. Nearest neighbor hot deck imputation has been used to generate fifty "complete" datasets. Adjusting for age and geographic location, the black/white mortality ratio for the imputed datasets was 3.3. When adjusted for all covariates, the mortality ratio was only 1.2. Overall, 87% of the excess mortality could be attributed to racial differences in disease stage, tumor characteristics, treatment, sociodemographic characteristics, hormonal and reproductive factors, the number of comorbidities and health behavior. The results based on multiple imputation indicate that missing data did not introduce major bias in the earlier analyses. However, multiple imputation provided narrower confidence intervals than those obtained previously. Multiple imputation was worthwhile since it gave more precise estimates for the relative mortality ratios. These findings have public health importance: they have implications for development of health policies and planning interventions to reduce the excess risk of death among black women with endometrial cancer.

Genesis is a feature-length science fiction/fantasy screenplay written for partial fulfillment of the requirements for the MFA degree in Imaging Arts/Film. The script tells the story of two parentless children and their rescuer as they journey across a barren landscape in a world disintegrating before an impending catastrophic event. Their goal is to reach a generation ship called Genesis designed to save only a portion of humankind. Along the way they encounter rebels and misfits who threaten and assist the three along their way. This submission includes: a thesis report which contains an analysis of the conception and writing process; the complete third draft script; the original proposal, and various early story-development documents.

Typically, firms make decisions in environment where both customer demand and also available capacity are uncertain. This research focuses on studying a firm's optimal response when both uncertainties are significant. In principle, to buffer against any variability, either safety stock of inventory or excess capacity is used. Since, however, there exists a broader range of alternatives that firms can choose to hedge their uncertainties. This dissertation systematically studies and compares different mechanisms that help firms to integrate operational improvements with the production and inventory control policies. The first essay aims to compare advance demand information and safety capacity option. We study the structure of the optimal production and safety capacity (e.g. outsourcing) decisions made by a manufacturer facing uncertain demand and capacity, who has some advance demand information. We then compare the conditions that make either of these two approaches more valuable for the manufacturer. Two of the most interesting findings are that (1) in general, conditions that make better demand information more valuable make the option to outsource less valuable (and vice versa) and (2) the combined effect of both options can be greater than the sum of the individual effects. It is thus able to provide managers with some guidance on how to think about their need for more safety capacity or better information (or both) when dealing with uncertainties. The second essay extends the above results and considers a firm with two plants having unreliable capacity. The firm's two alternatives include cooperative planning and transshipment, both of which help to reduce the uncertainty in demand and in capacity. The essay shows how the capacity uncertainty influences firms' production and transshipment decisions. Through a numerical study, we explore situations where implementing the optimal policy is most beneficial. The model allows us to evaluate the benefits of flexibility resulting from two sources of capacity combined with the benefits of transshipment, and to quantify some of the claims used in the collaborative planning programs. The remainder of the dissertation is dedicated to a study of decentralized system and focuses on the feasibility of coordination. The question is in the framework of decentralized distribution system with transshipment. We are interested in whether the two independent facilities can be induced to make the centralized decisions by a pair of predetermined transshipment prices.

This thesis focuses on the economic determinants of the occupational outmigration of labor from agriculture in the United States. An individual's decision to move out of the agricultural sector is primarily based on a comparison between the expected returns in the agricultural and nonagricultural sectors. Using annual time series data spanning the period from 1939 to 2004, an OLS regression model is estimated to evaluate the effects of economic and demographic variable on occupational out-migration from the agricultural sector. The determining factors investigated in the study are sectoral income differentials, off-farm income acquired by farm households, federal government payments to the agricultural sector, agricultural land values, and the unemployment rate in the U.S. economy at large. Three groups of agricultural labor migrants are considered: hired farm workers, farm operators, and total agricultural labor. The results suggest that high economic returns in a composite nonfarm sector relative to those attained in agriculture increase agricultural labor out migration. On the other hand, increased levels of off-farm income are associated with halting total agricultural labor out migration. Further, agricultural land values have a mixed effect on migration In particular, higher land values are found to decrease farm operator out migration, but increase hired farm worker out migration. Farm program payments were shown to be ineffective in staving off agricultural labor out migration. The unemployment rate in the economy as a whole did not significantly effect off-farm migration.