Measuring, Modeling, and Forecasting Volatility and Correlations from High-Frequency Data

This dissertation contains four essays that all share a common purpose: developing new methodologies to exploit the potential of high-frequency data for the measurement, modeling and forecasting of financial assets volatility and correlations. The first two chapters provide useful tools for univariate applications while the last two chapters develop multivariate methodologies. In chapter 1, we introduce a new class of univariate volatility models named FloGARCH models. FloGARCH models provide a parsimonious joint model for low frequency returns and realized measures, and are sufficiently flexible to capture long memory as well as asymmetries related to leverage effects. We analyze the performances of the models in a realistic numerical study and on the basis of a data set composed of 65 equities. Using more than 10 years of high-frequency transactions, we document significant statistical gains related to the FloGARCH models in terms of in-sample fit, out-of-sample fit and forecasting accuracy compared to classical and Realized GARCH models. In chapter 2, using 12 years of high-frequency transactions for 55 U.S. stocks, we argue that combining low-frequency exogenous economic indicators with high-frequency financial data improves the ability of conditionally heteroskedastic models to forecast the volatility of returns, their full multi-step ahead conditional distribution and the multi-period Value-at-Risk. Using a refined version of the Realized LGARCH model allowing for time-varying intercept and implemented with realized kernels, we document that nominal corporate profits and term spreads have strong long-run predictive ability and generate accurate risk measures forecasts over long-horizon. The results are based on several loss functions and tests, including the Model Confidence Set. Chapter 3 is a joint work with David Veredas. We study the class of disentangled realized estimators for the integrated covariance matrix of Brownian semimartingales with finite activity jumps. These estimators separate correlations and volatilities. We analyze different combinations of quantile- and median-based realized volatilities, and four estimators of realized correlations with three synchronization schemes. Their finite sample properties are studied under four data generating processes, in presence, or not, of microstructure noise, and under synchronous and asynchronous trading. The main finding is that the pre-averaged version of disentangled estimators based on Gaussian ranks (for the correlations) and median deviations (for the volatilities) provide a precise, computationally efficient, and easy alternative to measure integrated covariances on the basis of noisy and asynchronous prices. Along these lines, a minimum variance portfolio application shows the superiority of this disentangled realized estimator in terms of numerous performance metrics. Chapter 4 is co-authored with Niels S. Hansen, Asger Lunde and Kasper V. Olesen, all affiliated with CREATES at Aarhus University. We propose to use the Realized Beta GARCH model to exploit the potential of high-frequency data in commodity markets. The model produces high quality forecasts of pairwise correlations between commodities which can be used to construct a composite covariance matrix. We evaluate the quality of this matrix in a portfolio context and compare it to models used in the industry. We demonstrate significant economic gains in a realistic setting including short selling constraints and transaction costs.

Towards Proactive Context-aware Computing and Systems

A primary goal of context-aware systems is delivering the right information at the right place and right time to users in order to enable them to make effective decisions and improve their quality of life. There are three key requirements for achieving this goal: determining what information is relevant, personalizing it based on the users’ context (location, preferences, behavioral history etc.), and delivering it to them in a timely manner without an explicit request from them. These requirements create a paradigm that we term as “Proactive Context-aware Computing”. Most of the existing context-aware systems fulfill only a subset of these requirements. Many of these systems focus only on personalization of the requested information based on users’ current context. Moreover, they are often designed for specific domains. In addition, most of the existing systems are reactive - the users request for some information and the system delivers it to them. These systems are not proactive i.e. they cannot anticipate users’ intent and behavior and act proactively without an explicit request from them. In order to overcome these limitations, we need to conduct a deeper analysis and enhance our understanding of context-aware systems that are generic, universal, proactive and applicable to a wide variety of domains. To support this dissertation, we explore several directions. Clearly the most significant sources of information about users today are smartphones. A large amount of users’ context can be acquired through them and they can be used as an effective means to deliver information to users. In addition, social media such as Facebook, Flickr and Foursquare provide a rich and powerful platform to mine users’ interests, preferences and behavioral history. We employ the ubiquity of smartphones and the wealth of information available from social media to address the challenge of building proactive context-aware systems. We have implemented and evaluated a few approaches, including some as part of the Rover framework, to achieve the paradigm of Proactive Context-aware Computing. Rover is a context-aware research platform which has been evolving for the last 6 years. Since location is one of the most important context for users, we have developed ‘Locus’, an indoor localization, tracking and navigation system for multi-story buildings. Other important dimensions of users’ context include the activities that they are engaged in. To this end, we have developed ‘SenseMe’, a system that leverages the smartphone and its multiple sensors in order to perform multidimensional context and activity recognition for users. As part of the ‘SenseMe’ project, we also conducted an exploratory study of privacy, trust, risks and other concerns of users with smart phone based personal sensing systems and applications. To determine what information would be relevant to users’ situations, we have developed ‘TellMe’ - a system that employs a new, flexible and scalable approach based on Natural Language Processing techniques to perform bootstrapped discovery and ranking of relevant information in context-aware systems. In order to personalize the relevant information, we have also developed an algorithm and system for mining a broad range of users’ preferences from their social network profiles and activities. For recommending new information to the users based on their past behavior and context history (such as visited locations, activities and time), we have developed a recommender system and approach for performing multi-dimensional collaborative recommendations using tensor factorization. For timely delivery of personalized and relevant information, it is essential to anticipate and predict users’ behavior. To this end, we have developed a unified infrastructure, within the Rover framework, and implemented several novel approaches and algorithms that employ various contextual features and state of the art machine learning techniques for building diverse behavioral models of users. Examples of generated models include classifying users’ semantic places and mobility states, predicting their availability for accepting calls on smartphones and inferring their device charging behavior. Finally, to enable proactivity in context-aware systems, we have also developed a planning framework based on HTN planning. Together, these works provide a major push in the direction of proactive context-aware computing.

Structure learning of context-specific graphical models

Abstract The ultimate problem considered in this thesis is modeling a high-dimensional joint distribution over a set of discrete variables. For this purpose, we consider classes of context-specific graphical models and the main emphasis is on learning the structure of such models from data. Traditional graphical models compactly represent a joint distribution through a factorization justi ed by statements of conditional independence which are encoded by a graph structure. Context-speci c independence is a natural generalization of conditional independence that only holds in a certain context, speci ed by the conditioning variables. We introduce context-speci c generalizations of both Bayesian networks and Markov networks by including statements of context-specific independence which can be encoded as a part of the model structures. For the purpose of learning context-speci c model structures from data, we derive score functions, based on results from Bayesian statistics, by which the plausibility of a structure is assessed. To identify high-scoring structures, we construct stochastic and deterministic search algorithms designed to exploit the structural decomposition of our score functions. Numerical experiments on synthetic and real-world data show that the increased exibility of context-specific structures can more accurately emulate the dependence structure among the variables and thereby improve the predictive accuracy of the models.

Structural Time Series Modeling: an Application to Environmental Variables

A structural time series model is one which is set up in terms of components which have a direct interpretation. In this paper, the discussion focuses on the dynamic modeling procedure based on the state space approach (associated to the Kalman filter), in the context of surface water quality monitoring, in order to analyze and evaluate the temporal evolution of the environmental variables, and thus identify trends or possible changes in water quality (change point detection). The approach is applied to environmental time series: time series of surface water quality variables in a river basin. The statistical modeling procedure is applied to monthly values of physico- chemical variables measured in a network of 8 water monitoring sites over a 15-year period (1999-2014) in the River Ave hydrological basin located in the Northwest region of Portugal.

The antecedents of post-initial adoption behavior in a S-D logic context

Repercussions of innovation adoption and diffusion studies have long been imperative to the success of novel introductions. However, perceptions and deductions of current innovation understandings have been changing over time. The paradigm shift from the goods-dominant (G-D) logic to the service-dominant (S-D) logic potentially makes the distinction between product (goods) innovation and service innovation redundant as the S-D logic lens views all innovations as service innovations (Vargo and Lusch, 2004; 2008; Lusch and Nambisan, 2015). From this perspective, product innovations are in essence service innovations, as goods serve as mere distribution mechanisms to deliver service. Nonetheless, the transition to such a broadened and transcending view of service innovation necessitates concurrently a change in the underlying models used to investigate innovation and its subsequent adoption. The present research addresses this gap by engendering a novel model for the most crucial period of service diffusion within the S-D logic context – the post-initial adoption phase, which demarcates an individual’s behavior after the initial adoption decision of a service. As a wellfounded understanding of service diffusion and the complementary innovation adoption still lingers in its infancy, the current study develops a model based on interdisciplinary domains mapping. Here fore, knowledge of the relatively established viral source domain is mapped to the comparatively undetermined target domain of service innovation adoption. To assess the model and test the importance of the explanatory variables, survey data from 750 respondents of a bank in Northern Germany is scrutinized by means of Structural Equation Modeling (SEM). The findings reveal that the continuance intention of a customer, actual usage of the service and the customer influencer value all constitute important postinitial adoption behavior that have meaningful implications for a successful service adoption. Second, the four constructs customer influencer value, organizational commitment, perceived usefulness and service customization are evidenced to have a differential impact on a iv customer’s post-initial adoption behavior. Third, this study indicates that post-initial adoption behavior further underlies the influence of a user’s age and besides that is also provoked by the internal and external environments of service adoption. Finally, this research amalgamates the broad view of service innovation by Nambisan and Lusch (2015) with the findings ensuing this enquiry’s model to arrive at a framework that it both, generalizable and practically applicable. Implications for academia and practitioners are captured along with avenues for future research.

3D Block Modeling and Reserve Estimation of a Garnet Deposit

The purpose of this project is to develop a three-dimensional block model for a garnet deposit in the Alder Gulch, Madison County, Montana. Garnets occur in pre-Cambrian metamorphic Red Wash gneiss and similar rocks in the vicinity. This project seeks to model the percentage of garnet in a deposit called the Section 25 deposit using the Surpac software. Data available for this work are drillhole, trench and grab sample data obtained from previous exploration of the deposit. The creation of the block model involves validating the data, creating composites of assayed garnet percentages and conducting basic statistics on composites using Surpac statistical tools. Variogram analysis will be conducted on composites to quantify the continuity of the garnet mineralization. A three-dimensional block model will be created and filled with estimates of garnet percentage using different methods of reserve estimation and the results compared.

MODELING AND HARDWARE-IN-THE-LOOP SIMULATION OF POWER-SPLIT HYBRID ELECTRIC VEHICLES

Conventional vehicles are creating pollution problems, global warming and the extinction of high density fuels. To address these problems, automotive companies and universities are researching on hybrid electric vehicles where two different power devices are used to propel a vehicle. This research studies the development and testing of a dynamic model for Prius 2010 Hybrid Synergy Drive (HSD), a power-split device. The device was modeled and integrated with a hybrid vehicle model. To add an electric only mode for vehicle propulsion, the hybrid synergy drive was modified by adding a clutch to carrier 1. The performance of the integrated vehicle model was tested with UDDS drive cycle using rule-based control strategy. The dSPACE Hardware-In-the-Loop (HIL) simulator was used for HIL simulation test. The HIL simulation result shows that the integration of developed HSD dynamic model with a hybrid vehicle model was successful. The HSD model was able to split power and isolate engine speed from vehicle speed in hybrid mode.

DEVELOPMENT OF PRECIPITATION HARDENABLE AL-SC-ZR-HF QUATERNARY ALLOYS THROUGH THERMODYNAMIC MODELING, AND ROOM-TEMPERATURE AND ELEVATED TEMPERATURE HARDNESS

Aluminum alloyed with small atomic fractions of Sc, Zr, and Hf has been shown to exhibit high temperature microstructural stability that may improve high temperature mechanical behavior. These quaternary alloys were designed using thermodynamic modeling to increase the volume fraction of precipitated tri-aluminide phases to improve thermal stability. When aged during a multi-step, isochronal heat treatment, two compositions showed a secondary room-temperature hardness peak up to 700 MPa at 450°C. Elevated temperature hardness profiles also indicated an increase in hardness from 200-300°C, attributed to the precipitation of Al3Sc, however, no secondary hardness response was observed from the Al3Zr or Al3Hf phases in this alloy.

SYSTEM DYNAMICS MODELING AS A QUANTITATIVE-QUALITATIVE FRAMEWORK FOR SUSTAINABLE WATER RESOURCES MANAGEMENT: INSIGHTS FOR WATER QUALITY POLICY IN THE GREAT LAKES REGION

Early water resources modeling efforts were aimed mostly at representing hydrologic processes, but the need for interdisciplinary studies has led to increasing complexity and integration of environmental, social, and economic functions. The gradual shift from merely employing engineering-based simulation models to applying more holistic frameworks is an indicator of promising changes in the traditional paradigm for the application of water resources models, supporting more sustainable management decisions. This dissertation contributes to application of a quantitative-qualitative framework for sustainable water resources management using system dynamics simulation, as well as environmental systems analysis techniques to provide insights for water quality management in the Great Lakes basin. The traditional linear thinking paradigm lacks the mental and organizational framework for sustainable development trajectories, and may lead to quick-fix solutions that fail to address key drivers of water resources problems. To facilitate holistic analysis of water resources systems, systems thinking seeks to understand interactions among the subsystems. System dynamics provides a suitable framework for operationalizing systems thinking and its application to water resources problems by offering useful qualitative tools such as causal loop diagrams (CLD), stock-and-flow diagrams (SFD), and system archetypes. The approach provides a high-level quantitative-qualitative modeling framework for "big-picture" understanding of water resources systems, stakeholder participation, policy analysis, and strategic decision making. While quantitative modeling using extensive computer simulations and optimization is still very important and needed for policy screening, qualitative system dynamics models can improve understanding of general trends and the root causes of problems, and thus promote sustainable water resources decision making. Within the system dynamics framework, a growth and underinvestment (G&U) system archetype governing Lake Allegan's eutrophication problem was hypothesized to explain the system's problematic behavior and identify policy leverage points for mitigation. A system dynamics simulation model was developed to characterize the lake's recovery from its hypereutrophic state and assess a number of proposed total maximum daily load (TMDL) reduction policies, including phosphorus load reductions from point sources (PS) and non-point sources (NPS). It was shown that, for a TMDL plan to be effective, it should be considered a component of a continuous sustainability process, which considers the functionality of dynamic feedback relationships between socio-economic growth, land use change, and environmental conditions. Furthermore, a high-level simulation-optimization framework was developed to guide watershed scale BMP implementation in the Kalamazoo watershed. Agricultural BMPs should be given priority in the watershed in order to facilitate cost-efficient attainment of the Lake Allegan's TP concentration target. However, without adequate support policies, agricultural BMP implementation may adversely affect the agricultural producers. Results from a case study of the Maumee River basin show that coordinated BMP implementation across upstream and downstream watersheds can significantly improve cost efficiency of TP load abatement.

MODELING AND EXPERIMENTAL SETUP OF AN HCCI ENGINE

For the past three decades the automotive industry is facing two main conflicting challenges to improve fuel economy and meet emissions standards. This has driven the engineers and researchers around the world to develop engines and powertrain which can meet these two daunting challenges. Focusing on the internal combustion engines there are very few options to enhance their performance beyond the current standards without increasing the price considerably. The Homogeneous Charge Compression Ignition (HCCI) engine technology is one of the combustion techniques which has the potential to partially meet the current critical challenges including CAFE standards and stringent EPA emissions standards. HCCI works on very lean mixtures compared to current SI engines, resulting in very low combustion temperatures and ultra-low NOx emissions. These engines when controlled accurately result in ultra-low soot formation. On the other hand HCCI engines face a problem of high unburnt hydrocarbon and carbon monoxide emissions. This technology also faces acute combustion controls problem, which if not dealt properly with yields highly unfavorable operating conditions and exhaust emissions. This thesis contains two main parts. One part deals in developing an HCCI experimental setup and the other focusses on developing a grey box modelling technique to control HCCI exhaust gas emissions. The experimental part gives the complete details on modification made on the stock engine to run in HCCI mode. This part also comprises details and specifications of all the sensors, actuators and other auxiliary parts attached to the conventional SI engine in order to run and monitor the engine in SI mode and future SI-HCCI mode switching studies. In the latter part around 600 data points from two different HCCI setups for two different engines are studied. A grey-box model for emission prediction is developed. The grey box model is trained with the use of 75% data and the remaining data is used for validation purpose. An average of 70% increase in accuracy for predicting engine performance is found while using the grey-box over an empirical (black box) model during this study. The grey-box model provides a solution for the difficulty faced for real time control of an HCCI engine. The grey-box model in this thesis is the first study in literature to develop a control oriented model for predicting HCCI engine emissions for control.

PARAMETRIC STUDY OF REAXFF SIMULATION PARAMETERS FOR MOLECULAR DYNAMICS MODELING OF REACTIVE CARBON GASES

Abstract The development of innovative carbon-based materials can be greatly facilitated by molecular modeling techniques. Although the Reax Force Field (ReaxFF) can be used to simulate the chemical behavior of carbon-based systems, the simulation settings required for accurate predictions have not been fully explored. Using the ReaxFF, molecular dynamics (MD) simulations are used to simulate the chemical behavior of pure carbon and hydrocarbon reactive gases that are involved in the formation of carbon structures such as graphite, buckyballs, amorphous carbon, and carbon nanotubes. It is determined that the maximum simulation time step that can be used in MD simulations with the ReaxFF is dependent on the simulated temperature and selected parameter set, as are the predicted reaction rates. It is also determined that different carbon-based reactive gases react at different rates, and that the predicted equilibrium structures are generally the same for the different ReaxFF parameter sets, except in the case of the predicted formation of large graphitic structures with the Chenoweth parameter set under specific conditions.

MODELING OF TRANSFER PATH FOR DETERMINATION OF COMBUSTION AND NOISE METRICS ON DIESEL ENGINES

Determination of combustion metrics for a diesel engine has the potential of providing feedback for closed-loop combustion phasing control to meet current and upcoming emission and fuel consumption regulations. This thesis focused on the estimation of combustion metrics including start of combustion (SOC), crank angle location of 50% cumulative heat release (CA50), peak pressure crank angle location (PPCL), and peak pressure amplitude (PPA), peak apparent heat release rate crank angle location (PACL), mean absolute pressure error (MAPE), and peak apparent heat release rate amplitude (PAA). In-cylinder pressure has been used in the laboratory as the primary mechanism for characterization of combustion rates and more recently in-cylinder pressure has been used in series production vehicles for feedback control. However, the intrusive measurement with the in-cylinder pressure sensor is expensive and requires special mounting process and engine structure modification. As an alternative method, this work investigated block mounted accelerometers to estimate combustion metrics in a 9L I6 diesel engine. So the transfer path between the accelerometer signal and the in-cylinder pressure signal needs to be modeled. Depending on the transfer path, the in-cylinder pressure signal and the combustion metrics can be accurately estimated - recovered from accelerometer signals. The method and applicability for determining the transfer path is critical in utilizing an accelerometer(s) for feedback. Single-input single-output (SISO) frequency response function (FRF) is the most common transfer path model; however, it is shown here to have low robustness for varying engine operating conditions. This thesis examines mechanisms to improve the robustness of FRF for combustion metrics estimation. First, an adaptation process based on the particle swarm optimization algorithm was developed and added to the single-input single-output model. Second, a multiple-input single-output (MISO) FRF model coupled with principal component analysis and an offset compensation process was investigated and applied. Improvement of the FRF robustness was achieved based on these two approaches. Furthermore a neural network as a nonlinear model of the transfer path between the accelerometer signal and the apparent heat release rate was also investigated. Transfer path between the acoustical emissions and the in-cylinder pressure signal was also investigated in this dissertation on a high pressure common rail (HPCR) 1.9L TDI diesel engine. The acoustical emissions are an important factor in the powertrain development process. In this part of the research a transfer path was developed between the two and then used to predict the engine noise level with the measured in-cylinder pressure as the input. Three methods for transfer path modeling were applied and the method based on the cepstral smoothing technique led to the most accurate results with averaged estimation errors of 2 dBA and a root mean square error of 1.5dBA. Finally, a linear model for engine noise level estimation was proposed with the in-cylinder pressure signal and the engine speed as components.

FLOOD RISK ASSESSMENT UNDER HISTORICAL AND PREDICTED LAND USE CHANGE USING CONTINUOUS HYDROLOGIC MODELING

Current procedures for flood risk estimation assume flood distributions are stationary over time, meaning annual maximum flood (AMF) series are not affected by climatic variation, land use/land cover (LULC) change, or management practices. Thus, changes in LULC and climate are generally not accounted for in policy and design related to flood risk/control, and historical flood events are deemed representative of future flood risk. These assumptions need to be re-evaluated, however, as climate change and anthropogenic activities have been observed to have large impacts on flood risk in many areas. In particular, understanding the effects of LULC change is essential to the study and understanding of global environmental change and the consequent hydrologic responses. The research presented herein provides possible causation for observed nonstationarity in AMF series with respect to changes in LULC, as well as a means to assess the degree to which future LULC change will impact flood risk. Four watersheds in the Midwest, Northeastern, and Central United States were studied to determine flood risk associated with historical and future projected LULC change. Historical single framed aerial images dating back to the mid-1950s were used along with Geographic Information Systems (GIS) and remote sensing models (SPRING and ERDAS) to create historical land use maps. The Forecasting Scenarios of Future Land Use Change (FORE-SCE) model was applied to generate future LULC maps annually from 2006 to 2100 for the conterminous U.S. based on the four IPCC-SRES future emission scenario conditions. These land use maps were input into previously calibrated Soil and Water Assessment Tool (SWAT) models for two case study watersheds. In order to isolate effects of LULC change, the only variable parameter was the Runoff Curve Number associated with the land use layer. All simulations were run with daily climate data from 1978-1999, consistent with the 'base' model which employed the 1992 NLCD to represent 'current' conditions. Output daily maximum flows were converted to instantaneous AMF series and were subsequently modeled using a Log-Pearson Type 3 (LP3) distribution to evaluate flood risk. Analysis of the progression of LULC change over the historic period and associated SWAT outputs revealed that AMF magnitudes tend to increase over time in response to increasing degrees of urbanization. This is consistent with positive trends in the AMF series identified in previous studies, although there are difficulties identifying correlations between LULC change and identified change points due to large time gaps in the generated historical LULC maps, mainly caused by unavailability of sufficient quality historic aerial imagery. Similarly, increases in the mean and median AMF magnitude were observed in response to future LULC change projections, with the tails of the distributions remaining reasonably constant. FORE-SCE scenario A2 was found to have the most dramatic impact on AMF series, consistent with more extreme projections of population growth, demands for growing energy sources, agricultural land, and urban expansion, while AMF outputs based on scenario B2 showed little changes for the future as the focus is on environmental conservation and regional solutions to environmental issues.

THE EFFECTS OF MODELING INSTRUCTION ON STUDENT LEARNING OF A NEWTONIAN FORCE CONCEPT

This study explores the effects of modeling instruction on student learning in physics. Multiple representations grounded in physical contexts were employed by students to analyze the results of inquiry lab investigations. Class whiteboard discussions geared toward a class consensus following Socratic dialogue were implemented throughout the modeling cycle. Lab investigations designed to address student preconceptions related to Newton’s Third Law were implemented. Student achievement was measured based on normalized gains on the Force Concept Inventory. Normalized FCI gains achieved by students in this study were comparable to those achieved by students of other novice modelers. Physics students who had taken a modeling Intro to Physics course scored significantly higher on the FCI posttest than those who had not. The FCI results also provided insight into deeply rooted student preconceptions related to Newton’s Third Law. Implications for instruction and the design of lab investigations related to Newton’s Third Law are discussed.

MOLECULAR MODELING OF EPON 862-DETDA / CARBON COMPOSITES

The thermoset epoxy resin EPON 862, coupled with the DETDA hardening agent, are utilized as the polymer matrix component in many graphite (carbon fiber) composites. Because it is difficult to experimentally characterize the interfacial region, computational molecular modeling is a necessary tool for understanding the influence of the interfacial molecular structure on bulk-level material properties. The purpose of this research is to investigate the many possible variables that may influence the interfacial structure and the effect they will have on the mechanical behavior of the bulk level composite. Molecular models are established for EPON 862-DETDA polymer in the presence of a graphite surface. Material characteristics such as polymer mass-density, residual stresses, and molecular potential energy are investigated near the polymer/fiber interface. Because the exact degree of crosslinking in these thermoset systems is not known, many different crosslink densities (degrees of curing) are investigated. It is determined that a region exists near the carbon fiber surface in which the polymer mass density is different than that of the bulk mass density. These surface effects extend ~10 Å into the polymer from the center of the outermost graphite layer. Early simulations predict polymer residual stress levels to be higher near the graphite surface. It is also seen that the molecular potential energy in polymer atoms decreases with increasing crosslink density. New models are then established in order to investigate the interface between EPON 862-DETDA polymer and graphene nanoplatelets (GNPs) of various atomic thicknesses. Mechanical properties are extracted from the models using Molecular Dynamics techniques. These properties are then implemented into micromechanics software that utilizes the generalized method of cells to create representations of macro-scale composites. Micromechanics models are created representing GNP doped epoxy with varying number of graphene layers and interfacial polymer crosslink densities. The initial micromechanics results for the GNP doped epoxy are then taken to represent the matrix component and are re-run through the micromechanics software with the addition of a carbon fiber to simulate a GNP doped epoxy/carbon fiber composite. Micromechanics results agree well with experimental data, and indicate GNPs of 1 to 2 atomic layers to be highly favorable. The effect of oxygen bonded to the surface of the GNPs is lastly investigated. Molecular Models are created for systems with varying graphene atomic thickness, along with different amounts of oxygen species attached to them. Models are created for graphene containing hydroxyl groups only, epoxide groups only, and a combination of epoxide and hydroxyl groups. Results show models of oxidized graphene to decrease in both tensile and shear modulus. Attaching only epoxide groups gives the best results for mechanical properties, though pristine graphene is still favored.