Downloads & Free Reading Options - Results

In this project, we developed three new statistical models for characterizing the dynamic evolution of networks and detecting changes deviating from the normal evolving trajectory. Modeling the dynamics of evolving networks and detecting abnormal changes is of great interest in many domains. However, the fundamental methodological development in statistical modeling has been lacking. Through this project, the PI has not only significantly extended her research area and generated ample results, but also established industrial collaboration and introduced new educational materials. The PI was successfully promoted to Associate Professor with tenure, for which this grant served as one honorable recognition.

Intrinsic noise, the stochastic cell-to-cell fluctuations in mRNAs and proteins, has been observed and proved to play important roles in cellular systems. Due to the recent development in single-cell-level measurement technology, the studies on intrinsic noise are becoming increasingly popular among scholars. The chemical master equation (CME) has been used to model the evolutions of complex chemical and biological systems since 1940, and are often served as the standard tool for modeling intrinsic noise in gene regulation system. A CME-based model can capture the discrete, stochastic, and dynamical nature of gene regulation system, and may offer casual and physical explanation of the observed data at single-cell level. Nonetheless, the complexity of CME also pose serious challenge for researchers in proposing practical modeling and inference frameworks. In this article, we will review the existing works on the modelings and inference of intrinsic noise in gene regulation system within the framework of CME model. We will explore the principles in constructing a CME model for studying gene regulation system and discuss the popular approximations of CME. Then we will study the simulation simulation methods as well as the analytical and numerical approaches that can be used to obtain solution to a CME model. Finally we will summary the exiting statistical methods that can be used to infer the unknown parameters or structures in CME model using single-cell-level gene expression data.

In a recent work \cite{LiuJoladSchZia13}, we introduced dynamic networks with preferred degrees and presented simulation and analytic studies of a single, homogeneous system as well as two interacting networks. Here, we extend these studies to a wider range of parameter space, in a more systematic fashion. Though the interaction we introduced seems simple and intuitive, it produced dramatically different behavior in the single- and two-network systems. Specifically, partitioning the single network into two identical sectors, we find the cross-link distribution to be a sharply peaked Gaussian. In stark contrast, we find a very broad and flat plateau in the case of two interacting identical networks. A sound understanding of this phenomenon remains elusive. Exploring more asymmetric interacting networks, we discover a kind of `universal behavior' for systems in which the `introverts' (nodes with smaller preferred degree) are far outnumbered. Remarkably, an approximation scheme for their degree distribution can be formulated, leading to very successful predictions.

Statistical modeling of physical laws connects experiments with mathematical descriptions of natural phenomena. The modeling is based on the probability density of measured variables expressed by experimental data via a kernel estimator. As an objective kernel the scattering function determined by calibration of the instrument is introduced. This function provides for a new definition of experimental information and redundancy of experimentation in terms of information entropy. The redundancy increases with the number of experiments, while the experimental information converges to a value that describes the complexity of the data. The difference between the redundancy and the experimental information is proposed as the model cost function. From its minimum, a proper number of data in the model is estimated. As an optimal, nonparametric estimator of the relation between measured variables the conditional average extracted from the kernel estimator is proposed. The modeling is demonstrated on noisy chaotic data.

In this paper, we present an analytical approach in obtaining the probability density function (pdf) of the random decision variable Y, formed at the output of power-cubic all-optical nonlinear preprocessor followed by the photodetector. Our approach can be used to accurately evaluate the performance of ultrafast pulse detection in the presence of Gaussian noise. Through rigorous Monte-Carlo simulation, the accuracy of widely used Gaussian approximation of decision variable Y is refuted. However, in this paper we show that the so called Log-Pearson type-3 probability density function (LP3 pdf) is an excellent representation for the decision variable Y . Three distinguishable parameters of the LP3 pdf are obtained through analytical derivation of three moments of the decision variable Y . Furthermore, toward a more realistic model, in addition to ASE Gaussian noise, the effects of shot and thermal noises are also included. Finally, using the presented analytical approach, it is shown that power-cubic preprocessor outperforms its quadratic counterparts, i.e., Second Harmonic Generation (SHG) and Two Photon Absorption (TPA) devices, in high power regime where shot and thermal noises can be neglected.

608 Seiten :

608 Seiten :

In high-throughput genomics, large-scale designed experiments are becoming common, and analysis approaches based on highly multivariate regression and anova concepts are key tools. Shrinkage models of one form or another can provide comprehensive approaches to the problems of simultaneous inference that involve implicit multiple comparisons over the many, many parameters representing effects of design factors and covariates. We use such approaches here in a study of cardiovascular genomics. The primary experimental context concerns a carefully designed, and rich, gene expression study focused on gene-environment interactions, with the goals of identifying genes implicated in connection with disease states and known risk factors, and in generating expression signatures as proxies for such risk factors. A coupled exploratory analysis investigates cross-species extrapolation of gene expression signatures--how these mouse-model signatures translate to humans. The latter involves exploration of sparse latent factor analysis of human observational data and of how it relates to projected risk signatures derived in the animal models. The study also highlights a range of applied statistical and genomic data analysis issues, including model specification, computational questions and model-based correction of experimental artifacts in DNA microarray data.

This report describes the research performed during a three year study of computer reliability as affected by system activity. Data collected over an eight year period on three generations of computers was analyzed to determine what relationships exist between failures and activity. The activity included type of utilization, as well as amount of activity (or load) on the computer at time of failure. The failures included both software and hardware (transient, intermittent, and permanent). Another aspect of the research was an evaluation of the fault tolerance provided by the computer system as indicated by software error detection and recovery rates. A third major area of research was the development of a methodology for analysis of failure prediction data. Keywords: Reliability Modeling; Failure/Load Relationship; Failure Prediction; Fault Tolerance; Computer Utilization Rates; Error Detection; Error Correction.

This report describes the research performed during a three year study of computer reliability as affected by system activity. Data collected over an eight year period on three generations of computers was analyzed to determine what relationships exist between failures and activity. The activity included type of utilization, as well as amount of activity (or load) on the computer at time of failure. The failures included both software and hardware (transient, intermittent, and permanent). Another aspect of the research was an evaluation of the fault tolerance provided by the computer system as indicated by software error detection and recovery rates. A third major area of research was the development of a methodology for analysis of failure prediction data. Keywords: Reliability Modeling; Failure/Load Relationship; Failure Prediction; Fault Tolerance; Computer Utilization Rates; Error Detection; Error Correction.

Program testing is studied for maximizing the reliability of computer programs. Several formulas are derived to estimate the number of tests required to say that a program being tested is correct with a given probability. Strategies to select those test cases which minimize the probability of program error while keeping the number of tests fixed, are investigated. (author)

My project addresses statistical and stochastic problems in the following fields: Lagrangian prediction (1), Lagrangian data assimilation (2), development and validation of Lagrangian stochastic models (LSM) (3). The long range scientific objectives of this study comprise rigorous determining limits of predictability for Lagrangian motion in semi-enclosed seas, littoral zones, and straits on time scales of days and weeks, elaborating concrete prediction schemes, developing optimal Lagrangian data assimilation algorithms, and identification of multi particle stochastic models aimed at incorporating them to ocean circulation models (OCM).

This report describes a study performed under ARO sponsorship, addressing the investigation of a novel way to reduce complex and extensive oxidation reaction mechanisms containing hundreds of species to a much smaller number of progress variables, typically by a factor of ten. Because the results have been documented in several manuscripts, this final report is in the form of an Executive Summary succinctly describing the results and putting them in perspective with respect to existing oxidation mechanism reduction schemes. The manuscripts are individually listed as Appendices, and attached to this report.

In this paper we provide a comprehensive study of statistical inference in linear and allied models which exhibit some analytic perturbations in their design and covariance matrices. We also indicate a few potential applications. In the theory of perturbations of linear operators it has been known for a long time that the so-called ``singular perturbations'' can have a big impact on solutions of equations involving these operators even when their size is small. It appears that so far the question of whether such undesirable phenomena can also occur in statistical models and their solutions has not been formally studied. The models considered in this article arise in the context of nonlinear models where a single parameter accounts for the nonlinearity.

A fleet of 40 Landing Craft Air Cushion (LCAC) vessels is managed by Assault Craft Unit-5 located at Camp Pendleton, CA. LCACs are used to transport weapons systems, equipment, cargo, and personnel of the assault element of the Marine Air/Ground Task Force from ship to shore and across the beach. It is important that the Commanding Officer be able to forecast, with a reasonable degree of accuracy, the number of LCACs that will be available for tasking a fixed number of days in advance. As the LCAC fleet ages its time in maintenance increases, which in turn increases the uncertainty of achieving availability targets. This thesis examines factors that contribute to the availability of LCACs on a daily basis. Using logistic regression, a forecast model is developed from past data on availability and maintenance that has a prediction standard error of approximately two to three craft. The model can be used not only to forecast the number of available LCACs, but also to assist in planning and scheduling to better manage availability.

This is the second in a series of Naval Surface Warfare Center Dahlgren Division (NSWCDD) technical reports concerning software reliability. The first report, A Survey of Software Reliability Modeling and Estimation, NSWC TR 82-171, discusses various approaches advocated for reliability estimation; reviews various models proposed for this estimation process; provides model assumptions, estimates of reliability, and the precision of those estimates; and provides the data required for the models' implementation. Eight software reliability models were selected to form the basis of a library. This library also contains data edit, transformation, general statistics, and Goodness-of-Fit functions. The original Statistical Modeling and Estimation of Reliability Functions for Software (SMERFS) Library was described in the 1985 version of this report. The enhanced library, which now contains 11 models and model applicability analyses, is explained herein. The SMERFS User's Guide, NSWCDD TR 84-373, Revision 3, explains the execution of the new SMERFS driver.

MPEG video traffic is expected to cause several problems in ATM networks, both from performance and from architectural viewpoint. For the solution of these difficulties, appropriate video traffic models are needed. A detailed statistical analysis of newly generated long MPEG encoded video sequences is presented and the results are compared to those of existing data sets. Based on the results of the analysis, a layered modeling scheme for MPEG video traffic is suggested which will simplify the finding of appropriate models for many performance analysis techniques.

The long-term goal of this research is the development of accurate, fast, and advanced statistical and dynamical nonlinear models of ocean surface waves based on first physical principles. These wave models will improve both long-term ocean surface wave forecasts and the prediction of strongly coherent events, such as freak waves, tsunamis, and wave breaking. The objectives of this project are as follows: (1) finding a physically correct wind input term for the Hasselmann equation, (2) understanding the balance of source terms in the Hasselmann equation, (3) investigating the interaction of different scales on the ocean surface (sea and swell), and (4) developing new water surface analytical models and methods. The approach employed advanced analytical techniques, including Hamiltonian formalism, self-similar solutions, analytical solutions of integral equations, numerical methods for the solution of integral and pseudo-differential equations, and the comparison of analytical and numerical results with experimental data. We found the new wind input term through experimental, theoretical, and numerical approaches. We also found the following: (1) that S(sub nl) has a leading role in the energy balance of wind-driven seas within the Hasselmann equation; (2) global visual observations are a useful tool for discriminating sea swells and wind-driven seas; (3) a new method for solving the exact S(sub nl) in the energy balance equation; and (4) a new canonical equation for one-dimensional surface waves.

The original proposal for this research project states its goals as follows: The object of this research project is to continue a broad program of research whose aims are: to develop the quantile and density-quantile function formulation of statistical data analysis and modeling problems; to develop robust methods of data analysis and modeling; to develop density estimation methods; and to develop minimum distance methods and approximation theory methods. The author proposes to implement our theoretical research in algorithms and computer software which provide methods useable by researchers concerned with important scientific and social problems, and to discuss applications which illustrate the applicability of the methods. The approach to statistical reasoning that our research program is attempting to develop has reached a synthesis that warrants its own name; we propose the name FUN.STAT. The FUN.STAT domain of statistical data model identification and parameter estimation combines (1) density-quantile function signatures of distributions, (2) entropy and information measures, and (3) functional statistical inference.

The goal of this project is to improve our ability to predict the behavioral response of beaked whales to mid-frequency active (MFA) sonar, by making better use of data already collected, or being collected as part of other projects. We aim to construct and fit mathematical models of the diving behavior of beaked whales, and their response to MFA sonar. These models will be parameterized by fitting them simultaneously to three sources of data: (1) short-term, high fidelity tagging studies on individual whales (some of which comes from animals exposed to acoustic stimuli); (2) medium-term satellite tagging studies of individual whales (some of which we hope will come from data collected during navy exercises); and (3) long-term passive acoustic monitoring from bottom-mounted hydrophones (much of which comes from data collected during navy exercises). All data will come from the Atlantic Undersea Test and Evaluation Center (AUTEC), Bahamas, and the surrounding area. Hence our models and predictions will be directly applicable to animals in that area, although we hope they will be of more general relevance. Outputs of the model are designed to be compatible with risk evaluation and mitigation tools and models developed under other ONR initiatives, such as Effects of Sound on the Marine Environment (ESME) and Population Consequences of Acoustic Disturbance (PCADS). Hence, the model will: (1) predict the behavioral responses of individual beaked whales to MFA sonar; (2) provide sufficient information to assess the level of take that is likely as a result of sonar operations; (3) provide sufficient information to allow the energetic costs of disturbance by MFA to be estimated; (4) provide a modeling framework within which information concerning behavioral responses of beaked whales can be interpreted.

My project addresses statistical and stochastic problems in the following fields: Lagrangian prediction and Lagrangian data assimilation (1), estimating transport and mixing parameters from tracer observations (2), and ocean model validation (3). The long range scientific goals of this study comprise determining limits of predictability for Lagrangian motion in semi-enclosed seas and littoral zones on time scales of days and weeks, estimating mixing and transport parameters in the upper ocean to improve performance of numerical models , and constructing statistical tests for model validation based on realistic confidence intervals for estimated mean fields and appropriate quantitative misfit measures.

This report describes a research project which measured pilot response to seven control systems simulating different handling qualities, quantitatively evaluated and compared the systems based on these measurements, and compared the quantitative system evaluation based on measured pilot performance with a qualitative evaluation using the Cooper-Harper technique. The objective of the project is implementation of a methodology for system evaluation via pilot performance to complement the current evaluation technique based on subjective ratings by test pilots. Pilot performance is determined through analysis of objective dynamic measurements of pilot response typical of flight test environments. In short, the methodology specifies a general approach for condensing the typically huge mound of measured test data accumulated during flight simulation experiments into meaningful quantities for system evaluation. The key element in the methodology is statistical modeling of a law for pilot control. Statistical modeling of pilot control provides an assessment of pilot performance in terms of standard statistical estimation parameters. The methodology requires that this control model be used to compute control input in a closed loop tracking task; the accuracy of the control model in performing this task is an important measure of pilot performance relevant to system evaluation. In addition, these parameters computed from the dynamic measurements of pilot performance are shown to enhance understanding of the aspects of the handling qualities underlying subjective rating techniques such as Cooper-Harper.

Symposium T Objective: To showcase recent developments in statistical mechanical modeling of non-equilibrium stmctures and processes in materials (including film growth, solidification processes, and bulk defect dynamics).

A large set of quasi-classical, direct dynamics trajectory simulations were performed for decomposition of 1, 5-dinitrobiuret (DNB) over a temperature range from 4000 to 6000 K, aimed at providing insight into DNB decomposition mechanisms. The trajectories revealed various decomposition paths, and reproduced the products (including HNCO, N2O, NO2, NO, and water) observed in DNB pyrolysis experiments. Using trajectory results as a guide, structures of intermediate complexes and transition states that might be important for decomposition were determined using density function theory calculations. Rice-Ramsperger-Kassel-Marcus (RRKM) theory was then utilized to examine behaviors of the energized reactant and intermediates and to determine unimolecular rates for crossing the various transition states. According to RRKM predictions, the dominant initial decomposition path of energized DNB corresponds to elimination of HNNO2H via a concerted mechanism where the molecular decomposition is accompanied with intra-molecular H-atom transfer from the central nitrogen to the terminal nitro oxygen. Other important paths correspond to elimination of NO2 and H2NNO(pi). NO(2) elimination is a simple N-N bond scission process. Formation and elimination of nitramide is however dynamically complicated, requiring twisting a -NHNO2 group out of the molecular plane, followed by an intramolecular reaction to form nitramide before its elimination. These two paths become significant at temperature above 1500 K, accounting for greater than 17% of DNB decomposition at 2000 K. This work demonstrates that quasi-classical trajectory simulations, in conjunction with electronic structure and RRKM calculations, are able to extract mechanisms, kinetics, dynamics and product branching ratios for the decomposition of complex energetic molecules, and predict how they vary with decomposition temperature.

This is a final report on the research project entitled Multiple Time Series Modeling and Time Series Theoretic Statistical Methods for the period 1 April 1981 - 31 March 1983. (Author)

New product development using extrusion cooking, low cost raw materials to meet consumer preference for convenient products are increasingly becoming popular in the field of Food Engineering. The raw materials are usually complex biological materials that may undergo different transformations due to high temperature, pressure and shear in the extruder. This study investigated the effects of extrusion cooking conditions on the physical and functional characteristics of rice-cowpea breakfast gruel and modeled the relationship between the input and output variables for optimum product quality. Fifteen composite blends based on central composite rotatable design (CCRD) in a response surface methodology (RSM) were formulated and extruded at different extrusion conditions of barrel temperature, feed moisture content and rice-legume ratio in the feeds. Data generated from the analysis of the extruded products were fitted to second-order polynomial equation with optimal point located. Results showed significant (p≤0.05) variation in physical and functional characteristics of the extrudates studied. The optimal points for process conditions were obtained at 120 o C barrel temperature, 20% feed moisture and 24% feed cowpea composition respectively at desirability value of 0.907, for product quality of  126.14 (expansion index), 0.214g/cm 3  (bulk density), 6.83(water absorption index), 8.46 (water solubility index), 99.02% (dispersibility) and viscosity of 12.83Nm 2 . Fitted models indicated significant coefficients with adjusted R 2   of 94.70%, 88.00%, 99.70%, 99.10%, 95.90% and 83.00% respectively for expansion index, bulk density, water absorption and solubility indices, dispersibility and viscosity, while the lack-of-fit test p-values were 0.608, 0.485, 0.513, 0.505, 0.297 and 0.200 respectively. It was concluded that the fitted models could be used to adequately represent the experimental data in their natural state and when extrusion variables are carefully selected, rice-cowpea blends could produce quality gruel of optimum quality for breakfast or complimentary feeding.

We employ a statistical model of North Atlantic tropical cyclone (TC) tracks to investigate the relationship between sea-surface temperature (SST) and North American TC landfall rates. The track model is conditioned on summer SST in the tropical North Atlantic being in either the 19 hottest or the 19 coldest years in the period 1950-2005. For each conditioning many synthetic TCs are generated and landfall rates computed. Compared to direct analysis of historical landfall, the track model reduces the sampling error by projecting information from the entire basin onto the coast. There are 46% more TCs in hot years than cold in the model, which is highly significant compared to random sampling and corroborates well documented trends in North Atlantic TC number in recent decades. In the absence of other effects, this difference results in a significant increase in model landfall rates in hot years, uniform along the coast. Hot-cold differences in the geographic distribution of genesis and in TC propagation do not significantly alter the overall landfall-rate difference in the model, and the net landfall rate is 1/4.7 yr in hot years and 1/3.1 yr in cold years. SST influence on genesis site and propagation does modify the geographic distribution of landfall, however. The Yucatan suffers 3 times greater landfall rate in hot years than cold, while the U.S. mid-Atlantic coast exhibits no significant change. Landfall probabilities increase in hot years compared to all years in Florida, the U.S Gulf coast, the Mexican Gulf coast, and Yucatan with at least 95% confidence.

We employ a statistical model of North Atlantic tropical cyclone (TC) tracks to investigate the relationship between sea-surface temperature (SST) and North American TC landfall rates. The track model is conditioned on summer SST in the tropical North Atlantic being in either the 19 hottest or the 19 coldest years in the period 1950-2005. For each conditioning many synthetic TCs are generated and landfall rates computed. Compared to direct analysis of historical landfall, the track model reduces the sampling error by projecting information from the entire basin onto the coast. There are 46% more TCs in hot years than cold in the model, which is highly significant compared to random sampling and corroborates well documented trends in North Atlantic TC number in recent decades. In the absence of other effects, this difference results in a significant increase in model landfall rates in hot years, uniform along the coast. Hot-cold differences in the geographic distribution of genesis and in TC propagation do not significantly alter the overall landfall-rate difference in the model, and the net landfall rate is 1/4.7 yr in hot years and 1/3.1 yr in cold years. SST influence on genesis site and propagation does modify the geographic distribution of landfall, however. The Yucatan suffers 3 times greater landfall rate in hot years than cold, while the U.S. mid-Atlantic coast exhibits no significant change. Landfall probabilities increase in hot years compared to all years in Florida, the U.S Gulf coast, the Mexican Gulf coast, and Yucatan with at least 95% confidence.

Fault detection, control, and forecasting have a vital role in renewable energy systems (Photovoltaics (PV) and wind turbines (WTs)) to improve their productivity, ef?ciency, and safety, and to avoid expensive maintenance. For instance, the main crucial and challenging issue in solar and wind energy production is the volatility of intermittent power generation due mainly to weather conditions. This fact usually limits the integration of PV systems and WTs into the power grid. Hence, accurately forecasting power generation in PV and WTs is of great importance for daily/hourly efficient management of power grid production, delivery, and storage, as well as for decision-making on the energy market. Also, accurate and prompt fault detection and diagnosis strategies are required to improve efficiencies of renewable energy systems, avoid the high cost of maintenance, and reduce risks of fire hazards, which could affect both personnel and installed equipment. This book intends to provide the reader with advanced statistical modeling, forecasting, and fault detection techniques in renewable energy systems.

3 v

This thesis researches methods of generating accurate and realistic polygonal terrain models by reducing gridded sampled terrain elevation data such as DMA DTED. The terrain models generated should be applicable for use in flight simulators and other systems. Existing methods of terrain modeling are discussed, and limitations of these systems are presented. The geostatistical estimation technique known as kriging is presented as a method estimating terrain elevations at locations not provided in DMA DTED or other gridded terrain elevation data. Kriging is an optimal interpolation method based on statistical analysis of the data. It also provides a measure of accuracy, the error variance, that allows some control over the accuracy of the resulting terrain model. This estimation method is employed in a terrain modeling system that builds polygonal terrain models at any resolution. This system is described in this thesis, and the results of terrain modeled both by filtering the DMA DTED and by estimating elevations with kriging are presented. The technique as implemented is computationally very expensive and has therefore limited the results of this thesis effort. Also, terrain models that could be produced appeared smoother than their filtered counterparts. However, kriging still shows promise as a method of estimating terrain elevations for terrain models.

This report discusses the application of concepts of statistical language theory (Zipf's Laws) to the derivation of formulas for measuring program and language complexity. Experimental data from several different programs and programming languages, such as PL/I, assembly and FORTRAN, is presented which is used to verify the necessary underlying assumption and to verify formulas for program length by comparison with actual statistics. Finally, the derived formulas are compared with those of Software Physics derived by Halstead. (Author)

This paper considers the application of system identification techniques using spectral representation for fitting models to textures and images and consists of two parts. In part I, we develop consistent decision rules for choosing the neighborhood in a one-dimensional autoregressive (AR) model. In part II, the theory is extended to the case of stationary two-dimensional random fields. (Author)

This article is from Molecular & Cellular Proteomics : MCP , volume 12 . Abstract Multiplexed bead-based flow cytometric immunoassays are a powerful experimental tool for investigating cellular communication networks, yet their widespread adoption is limited in part by challenges in robust quantitative analysis of the measurements. Here we report our application of mixed-effects modeling for the normalization and statistical analysis of bead-based immunoassay data. Our data set consisted of bead-based immunoassay measurements of 16 phospho-proteins in lysates of HepG2 cells treated with ligands that regulate acute-phase protein secretion. Mixed-effects modeling provided estimates for the effects of both the technical and biological sources of variance, and normalization was achieved by subtracting the technical effects from the measured values. This approach allowed us to detect ligand effects on signaling with greater precision and sensitivity and to more accurately characterize the HepG2 cell signaling network using constrained fuzzy logic. Mixed-effects modeling analysis of our data was vital for ascertaining that IL-1α and TGF-α treatment increased the activities of more pathways than IL-6 and TNF-α and that TGF-α and TNF-α increased p38 MAPK and c-Jun N-terminal kinase (JNK) phospho-protein levels in a synergistic manner. Moreover, we used mixed-effects modeling-based technical effect estimates to reveal the substantial variance contributed by batch effects along with the absence of loading order and assay plate position effects. We conclude that mixed-effects modeling enabled additional insights to be gained from our data than would otherwise be possible and we discuss how this methodology can play an important role in enhancing the value of experiments employing multiplexed bead-based immunoassays.

This article is from Molecular & Cellular Proteomics : MCP , volume 12 . Abstract Multiplexed bead-based flow cytometric immunoassays are a powerful experimental tool for investigating cellular communication networks, yet their widespread adoption is limited in part by challenges in robust quantitative analysis of the measurements. Here we report our application of mixed-effects modeling for the normalization and statistical analysis of bead-based immunoassay data. Our data set consisted of bead-based immunoassay measurements of 16 phospho-proteins in lysates of HepG2 cells treated with ligands that regulate acute-phase protein secretion. Mixed-effects modeling provided estimates for the effects of both the technical and biological sources of variance, and normalization was achieved by subtracting the technical effects from the measured values. This approach allowed us to detect ligand effects on signaling with greater precision and sensitivity and to more accurately characterize the HepG2 cell signaling network using constrained fuzzy logic. Mixed-effects modeling analysis of our data was vital for ascertaining that IL-1α and TGF-α treatment increased the activities of more pathways than IL-6 and TNF-α and that TGF-α and TNF-α increased p38 MAPK and c-Jun N-terminal kinase (JNK) phospho-protein levels in a synergistic manner. Moreover, we used mixed-effects modeling-based technical effect estimates to reveal the substantial variance contributed by batch effects along with the absence of loading order and assay plate position effects. We conclude that mixed-effects modeling enabled additional insights to be gained from our data than would otherwise be possible and we discuss how this methodology can play an important role in enhancing the value of experiments employing multiplexed bead-based immunoassays.

A didactic text with 3 figures.

In this report we summarize the research effort, funded under ONR Grant N00014-91-J-1004, on probabilistic modeling and statistical inference for random fields and space-time processes. The research we have pursued in this project consists of the investigation of a set of interrelated topics involving the development of new mathematical methods for the challenging problems of random field analysis and inference and the application of these methods to problems of practical significance. In particular, the problems of interest in this work were motivated by and directly address issues that are central both to the challenging large-scale data assimilation and estimation problems arising in physical oceanography and to a number of other remote sensing and imaging problems of direct interest to the Navy. (AN)

The purpose of this thesis is to integrate statistical/quantitative background material with Master's level analytical courses. This thesis first identifies the requirements for management education in terms of AACSB and NASPAA standards. Then, based on a comparative analysis of the country's top master's of business administration (MBA) programs and Naval Postgraduate School's current Systems Management curricula, and a survey conducted among SM faculty members, it integrates the decision-making modeling with statistical/quantitative background material for master's level analytical courses. The structure of the MS in Management at NPS, while satisfying the requirements of both AACSB and NASPAA is similar to the top management schools' MBA programs in the United States. However, top management schools' statistical/quantitative course sequence generally has four courses, providing more statistic al/quantitative background material than those three of NPS. Additionally, the contents of these three courses are not offered in adequate depth and some topics are duplicated. The new sequence and the contents of these courses are proposed based on a survey conducted among SM faculty members.

This report details progress on a joint project between Duke University, the University of British Columbia and Sky Research, on the development of advanced statistical signal processing and geophysical modeling tools for sensing buried unexploded ordnance (UXO). The research reported here discusses feature extraction, workflow development, and exploitation of additional knowledge about UXO (that they are rare). Further, we have performed a detailed comparison of different classification strategies, across the different components of the research team. The methods are tested throughout using measured data from actual sites, often utilizing data collected under related ESTCP projects.

The objective of the project, to date, has been to focus on biological issues believed to be important in converting toxin dosage to pre- cancerous and cancerous cells, and to translate these into appropriate quantitative mathematical terms. In particular, models that stem from the clonal expansion mechanisms identified by Moolgavkar and co-workers, cf, sample references (1979), (1983), have been studied, generalized (to account for possible variability or susceptibility between individual fish), and adapted for fitting to actual data. It appears to be widely believed that pre-cancerous conditions in an organ (the liver) occur as a result of cell clonal expansion, followed by a promotion (to tumor) event. Specific models for this has been proposed and developed by Moolgavkar and co-workers. More recent work is by C. J. Portier and co-workers.

Validating a concept of operation for a complex, safety-critical system (like the National Airspace System) is challenging because of the high dimensionality of the controllable parameters and the infinite number of states of the system. In this paper, we use statistical modeling techniques to explore the behavior of a conflict detection and resolution algorithm designed for the terminal airspace. These techniques predict the robustness of the system simulation to both nominal and off-nominal behaviors within the overall airspace. They also can be used to evaluate the output of the simulation against recorded airspace data. Additionally, the techniques carry with them a mathematical value of the worth of each prediction-a statistical uncertainty for any robustness estimate. Uncertainty Quantification (UQ) is the process of quantitative characterization and ultimately a reduction of uncertainties in complex systems. UQ is important for understanding the influence of uncertainties on the behavior of a system and therefore is valuable for design, analysis, and verification and validation. In this paper, we apply advanced statistical modeling methodologies and techniques on an advanced air traffic management system, namely the Terminal Tactical Separation Assured Flight Environment (T-TSAFE). We show initial results for a parameter analysis and safety boundary (envelope) detection in the high-dimensional parameter space. For our boundary analysis, we developed a new sequential approach based upon the design of computer experiments, allowing us to incorporate knowledge from domain experts into our modeling and to determine the most likely boundary shapes and its parameters. We carried out the analysis on system parameters and describe an initial approach that will allow us to include time-series inputs, such as the radar track data, into the analysis

This article is from Frontiers in Human Neuroscience , volume 7 . Abstract Valid research on neglect rehabilitation demands a statistical approach commensurate with the characteristics of neglect rehabilitation data: neglect arises from impairment in distinct brain networks leading to large between-subject variability in baseline symptoms and recovery trajectories. Studies enrolling medically ill, disabled patients, may suffer from missing, unbalanced data, and small sample sizes. Finally, assessment of rehabilitation requires a description of continuous recovery trajectories. Unfortunately, the statistical method currently employed in most studies of neglect treatment [repeated measures analysis of variance (ANOVA), rANOVA] does not well-address these issues. Here we review an alternative, mixed linear modeling (MLM), that is more appropriate for assessing change over time. MLM better accounts for between-subject heterogeneity in baseline neglect severity and in recovery trajectory. MLM does not require complete or balanced data, nor does it make strict assumptions regarding the data structure. Furthermore, because MLM better models between-subject heterogeneity it often results in increased power to observe treatment effects with smaller samples. After reviewing current practices in the field, and the assumptions of rANOVA, we provide an introduction to MLM. We review its assumptions, uses, advantages, and disadvantages. Using real and simulated data, we illustrate how MLM may improve the ability to detect effects of treatment over ANOVA, particularly with the small samples typical of neglect research. Furthermore, our simulation analyses result in recommendations for the design of future rehabilitation studies. Because between-subject heterogeneity is one important reason why studies of neglect treatments often yield conflicting results, employing statistical procedures that model this heterogeneity more accurately will increase the efficiency of our efforts to find treatments to improve the lives of individuals with neglect.

Diagnostic model output parameters, provided by the Fleet Numerical Weather Central, Monterey, California (FNWC) , and the marine fog frequency climatology developed at the Naval Postgraduate School, Monterey, California, are statistically processed in context with marine surface synoptic ship reports for the purpose of developing a linear regression scheme for modeling the distribution of marine fog. The study area encompasses a large section of the North Pacific Ocean (from 30-60N) . The time period involves 0000 GMT, 1-30 July 1976. The predictand is a fog parameter developed from a quantitative categorization of each of the 4481 synoptic observations according to reported present and past weather, low-cloud type, and visibility codes. The 38 model output and climatological predictor parameters are interpolated to each of the ship observation positions and the resultant data file is used for the derivation of the regression equations. The diagnostic capabilities of the regression equations, along with other existent approaches, are analyzed through the use of three verification scoring systems--Heidke Skill, Threat, and the Panofsky-Brier Probability scores. Improvement over climatology and FNWC's operational fog probability program (FTER, applied at analysis time) is demonstrated. Selective mappings of the regression equation outputs and categorized observations are intercompared with the sea-level pressure analysis; FTER; and the most significant predictor parameter in the regression equations, evaporative heat flux.

Diagnostic model output parameters, provided by the Fleet Numerical Weather Central, Monterey, California (FNWC) , and the marine fog frequency climatology developed at the Naval Postgraduate School, Monterey, California, are statistically processed in context with marine surface synoptic ship reports for the purpose of developing a linear regression scheme for modeling the distribution of marine fog. The study area encompasses a large section of the North Pacific Ocean (from 30-60N) . The time period involves 0000 GMT, 1-30 July 1976. The predictand is a fog parameter developed from a quantitative categorization of each of the 4481 synoptic observations according to reported present and past weather, low-cloud type, and visibility codes. The 38 model output and climatological predictor parameters are interpolated to each of the ship observation positions and the resultant data file is used for the derivation of the regression equations. The diagnostic capabilities of the regression equations, along with other existent approaches, are analyzed through the use of three verification scoring systems--Heidke Skill, Threat, and the Panofsky-Brier Probability scores. Improvement over climatology and FNWC's operational fog probability program (FTER, applied at analysis time) is demonstrated. Selective mappings of the regression equation outputs and categorized observations are intercompared with the sea-level pressure analysis; FTER; and the most significant predictor parameter in the regression equations, evaporative heat flux.

Bibliography: l. 111-112