Physical Sciences and Mathematics Commons^™

Ensemble Data Fitting For Bathymetric Models Informed By Nominal Data, Samantha Zambo

Dissertations

Due to the difficulty and expense of collecting bathymetric data, modeling is the primary tool to produce detailed maps of the ocean floor. Current modeling practices typically utilize only one interpolator; the industry standard is splines-in-tension.

In this dissertation we introduce a new nominal-informed ensemble interpolator designed to improve modeling accuracy in regions of sparse data. The method is guided by a priori domain knowledge provided by artificially intelligent classifiers. We recast such geomorphological classifications, such as ‘seamount’ or ‘ridge’, as nominal data which we utilize as foundational shapes in an expanded ordinary least squares regression-based algorithm. To our knowledge …

A Management Strategy Evaluation Of The Impacts Of Interspecific Competition And Recreational Fishery Dynamics On Vermilion Snapper (Rhomboplites Aurorubens) In The Gulf Of Mexico, Megumi C. Oshima

Dissertations

In the Gulf of Mexico (GOM), Vermilion Snapper (Rhomboplites auroruben), are believed to compete with Red Snapper directly for prey and habitat. The two species share similar diets and have significant spatial overlap in the Gulf. Red Snapper are thought to be the dominate competitor, forcing Vermilion Snapper to feed on less nutritious prey when local resources are depleted. In addition to ecological pressures, GOM Vermilion Snapper support substantial commercial and recreational fisheries. Over the past decade, recreational landings have steadily increased, reaching a historical high in 2018. One cause may be stricter regulations for similar target species such as …

Machine Learning Approaches For Improving Prediction Performance Of Structure-Activity Relationship Models, Gabriel Idakwo

Dissertations

In silico bioactivity prediction studies are designed to complement in vivo and in vitro efforts to assess the activity and properties of small molecules. In silico methods such as Quantitative Structure-Activity/Property Relationship (QSAR) are used to correlate the structure of a molecule to its biological property in drug design and toxicological studies. In this body of work, I started with two in-depth reviews into the application of machine learning based approaches and feature reduction methods to QSAR, and then investigated solutions to three common challenges faced in machine learning based QSAR studies.

First, to improve the prediction accuracy of learning …

The Influence Of The Electric Supply Industry On Economic Growth In Less Developed Countries, Edward Richard Bee

Dissertations

This study measures the impact that electrical outages have on manufacturing production in 135 less developed countries using stochastic frontier analysis and data from World Bank’s Investment Climate surveys. Outages of electricity, for firms with and without backup power sources, are the most frequently cited constraint on manufacturing growth in these surveys.

Outages are shown to reduce output below the production frontier by almost five percent in Africa and by a lower percentage in South Asia, Southeast Asia and the Middle East and North Africa. Production response to outages is quadratic in form. Outages also increase labor cost, reduce exports …

Comparison Of Hazard, Odds And Risk Ratio In The Two-Sample Survival Problem, Benedict P. Dormitorio

Dissertations

Cox proportional hazards is the standard method for analyzing treatment efficacy when time-to-event data is available. In the absence of time-to-event, investigators may use logistic regression which only requires relative frequencies of events, or Poisson regression which requires only interval-summarized frequency tables of time-to-event. When event frequencies are used instead of time-to-events, does it always result in a loss in power?

We investigate the relative performance of the three methods. In particular, we compare the power of tests based on the respective effect-size estimates (1)hazard ratio (HR), (2)odds ratio (OR), and (3)risk ratio (RR). We use a variety of survival …

Estimation And Inference For Spatial And Spatio-Temporal Mixed Effects Models, Casey M. Jelsema

Dissertations

One of the most common goals of geostatistical analysis is that of spatial prediction, in other words: filling in the blank areas of the map. There are two popular methods for accomplishing spatial prediction. Either kriging, or Bayesian hierarchical models. Both methods require the inverse of the spatial covariance matrix of the data. As the sample size, n, becomes large, both of these methods become impractical. Reduced rank spatial models (RRSM) allow prediction on massive datasets without compromising the complexity of the spatial process. This dissertation focuses on RRSMs, particularly situations where the data follow non-Gaussian distributions.

The manner in …

A Robust Estimate For The Bifurcating Autoregressive Model With Application To Cell Lineage Data, Tamer M. E. Elbayoumi

Dissertations

The bifurcating autoregressive model (BAR) is commonly used to model binary tree data. One application for this model relates to cell lineage data in biology. The purpose of studying the cell lineage process is to know whether the observed correlations between related cells are due to similarities in the environmental, inherited effects, or a combination of both of them. Because outliers in this kind of data are quite common, the need for a robust estimation procedure is necessary. A weighted L1 (WL1) estimate for estimating the parameters of the BAR model is considered. When the weights are constant, the estimate …

Rank-Based Estimation And Prediction For Mixed Effects Models In Nested Designs, Yusuf K. Bilgic

Dissertations

Hierarchical designs frequently occur in many research areas. The experimental design of interest is expressed in terms of fixed effects but, for these designs, nested factors are a natural part of the experiment. These nested effects are generally considered random and must be taken into account in the statistical analysis. Traditional analyses are quite sensitive to outliers and lose considerable power to detect the fixed effects of interest.

This work proposes three rank-based fitting methods for handling random, fixed and scale effects in k-level nested designs for estimation and inference. An algorithm, which iteratively obtains robust prediction for both scale …

Robust Residuals And Diagnostics In Autoregressive Time Series, Kirk W. Anderson

Dissertations

One of the goals of model diagnostics is outlier detection. In particular, we would like to use the residuals, appropriately standardized, to “flag” outliers. Hopefully, our (robust) procedure has yielded a fit that resists undue influence by outlying points, while simultaneously drawing attention to these interesting points via residual analysis. In this study we consider several different methods of standardizing the residuals resulting from autoregression. A large sample approximation for the variance of rank-based first order autoregressive time series residuals is developed. This provides studentized residuals, specific to the time series model and estimation procedure. Simulation studies are presented that …

New Statitstical Methods For The Estimation Of The Mean And Standard Deviation From Normally Distributed Censored Samples, Abou El-Makarim Abd El-Alim Aboueissa

Dissertations

The main objective of this dissertation is to estimate the mean /x and standard deviation cr of a normal population from left-censored samples. We have developed new methods for calculating estimates for the mean and standard deviation of a normal population from left-censored samples. Some of these methods based on traditional estimating procedures. A new method of obtaining the Cohen maximum likelihood estimates for fx and cr without the aid of an auxiliary table will be introduced. This new method will be used to extend Cohen table of estimating the Cohen A-parameter that is required for calculating the maximum likelihood …

On Rank-Based Considerations For Generalized Linear Models And Generalized Estimating Equation Models, Diana R. Cucos

Dissertations

This study discusses rank-based robust methods for estimation of parameters and hypotheses testing in the generalized linear models (GLM) and generalized estimating equations (GEE) setting. The robust estimates are obtained by minimizing a Wilcoxon drop in dispersion function for linear or nonlinear regression models. In addition, diagnostic tools for outliers and influential observations are being developed. These models are generalizations of linear and nonlinear models. They allow for both nonlinear mean functions and heteroscedasticity of their random errors. This makes them quite useful in practice. Rank-based inference has been developed for linear models over the last thirty years. This inference …

Visualization Methods: A Comparative Study Of New, Traditional And Robust Procedures, Kimberly Crimin

Dissertations

Two major goals in discriminant analysis are discrimination and classification. In discrimination, the goal is to describe graphically (visualization) different features of several known groups. In classification, the goal is to allocate unknown observations to one of several known groups. We have developed new visualization procedures based on traditional estimating procedures and also on robust estimating procedures. We have further developed robust classification procedures. We propose several robust classification procedures based on coordinatewise and affine equivariant, rank-based robust estimates. Empirical studies are performed over many different error distributions. These studies result in empirical efficiencies of the robust and traditional procedures. …

Nonlinear Regression Based On Ranks, Ashebar Abebe

Dissertations

This study presents robust methods for estimating parameters of nonlinear regression models. The proposed methods obtain estimates by minimizing rankbased dispersions instead of the Euclidean norm. We focus on the Wilcoxon and generalized signed-rank dispersion functions. Asymptotic properties of the estimators are established under mild regularity conditions similar to those used in least squares and least absolute deviations estimation. The study also shows that by considering the generalized signed-rank dispersion we obtain a class of estimators that encompasses most of the existing popular nonlinear regression estimators. As in linear models, these rank-based procedures provide estimators that are highly efficient. This …