This thesis looks at the problem of developmental respiration in Sarcophaga crassipalpis Macquart from the biological and instrumental points of view and adapts mathematical and statistical tools in order to analyze the data gathered. The biological motivation and current state of research is given as well as instrumental considerations and problems in the measurement of carbon dioxide production. A wide set of mathematical and statistical tools are used to analyze the time series produced in the laboratory. The objective is to assemble a methodology for the production and analysis of data that can be used in further developmental respiration research.

To be effective, a linear congruential random number generator (LCG) should produce values that are (a) uniformly distributed on the unit interval (0,1) excluding endpoints and (b) substantially free of serial correlation. It has been found that many statistical methods produce inflated Type I error rates for correlated observations. Theoretically, independently seeding an LCG under the following conditions attenuates serial correlation: (a) simple random sampling of seeds, (b) non-replicate streams, (c) non-overlapping streams, and (d) non-adjoining streams. Accordingly, 4 algorithms (each satisfying at least 1 condition) were developed: (a) zero-leap, (b) fixed-leap, (c) scaled random-leap, and (d) unscaled random-leap. Note that the latter satisfied all 4 independent seeding conditions. To assess serial correlation, univariate and multivariate simulations were conducted at 3 equally spaced intervals for each algorithm (N=24) and measured using 3 randomness tests: (a) the serial correlation test, (b) the runs up test, and (c) the white noise test. A one-way balanced multivariate analysis of variance (MANOVA) was used to test 4 hypotheses: (a) omnibus, (b) contrast of unscaled vs. others, (c) contrast of scaled vs. others, and (d) contrast of fixed vs. others. The MANOVA assumptions of independence, normality, and homogeneity were satisfied. In sum, the seeding algorithms did not differ significantly from each other (omnibus hypothesis). For the contrast hypotheses, only the fixed-leap algorithm differed significantly from all other algorithms. Surprisingly, the scaled random-leap offered the least difference among the algorithms (theoretically this algorithm should have produced the second largest difference). Although not fully supported by the research design used in this study, it is thought that the unscaled random-leap algorithm is the best choice for independently seeding the multiplicative congruential random number generator. Accordingly, suggestions for further research are proposed.

Matrix variate skew elliptically contoured distributions generalize several classes of important distributions. This paper defines and explores matrix variate skew elliptically contoured distributions. In particular, we discuss two stochastic representations of the matrix variate skew elliptically contoured distributions.

Given the rise in the application of neural networks to all sorts of interesting problems, it seems natural to apply them to statistical tests. This senior thesis studies whether neural networks built to classify discrete circular probability distributions can outperform a class of well-known statistical tests for uniformity for discrete circular data that includes the Rayleigh Test1, the Watson Test2, and the Ajne Test3. Each neural network used is relatively small with no more than 3 layers: an input layer taking in discrete data sets on a circle, a hidden layer, and an output layer outputting probability values between 0 and 1, with 0 mapping to uniform and 1 mapping to nonuniform. In evaluating performances, I compare the accuracy, type I error, and type II error of this class of statistical tests and of the neural networks built to compete with them. 1 Jammalamadaka, S. Rao(1-UCSB-PB); SenGupta, A.(6-ISI-ASU)Topics in circular statistics. (English summary) With 1 IBM-PC floppy disk (3.5 inch; HD). Series on Multivariate Analysis, 5. World Scientific Publishing Co., Inc., River Edge, NJ, 2001. xii+322 pp. ISBN: 981-02-3778-2 2 Watson, G. S.Goodness-of-fit tests on a circle. II. Biometrika 49 1962 57–63. 3 Ajne, B.A simple test for uniformity of a circular distribution. Biometrika 55 1968 343–354.

In 1993, a travel study was conducted along the Wasatch front in Utah (Research Systems Group INC, 2013). The main purpose of this study was to assess travel behavior to understand the needs for future growth in Utah. Since then, the Research Service Group (RSG), conducted a new study in 2012 to understand current travel preferences in Utah. This survey, called the Residential Choice Stated Preference survey, asked respondents to make ten choice comparisons between two hypothetical homes. Each home in the choice comparison was described by different attributes, those attributes that were used are, type of neighborhood, distance from important destinations, distance from access to public transport, street design, parking availability, commute distance to work, and price. The survey was designed to determine the extent to which Utah residents prefer alternative household attributes in a choice selection. Each attribute contained multiple characteristic levels that were randomly combined to define each alternative home in each choice comparison. Those choices can be explained by Random Utility Theory. Multinomial logistic regression will be used to estimate changes in utility when alternative attribute levels are present in a choice comparison. Using the coefficient estimate for price, a marginal willingness to pay (MWTP) for each attribute level will be calculated. This paper will use two different approaches to obtain MWTP estimates. Method One will use housing and rent price to recode the price variable in dollar terms as defined in the discrete choice experiment. Method Two will recode the price variable as an average ten percent change in home value to extrapolate a one-time payment for homes. As a result, we found that it is possible to obtain willingness to pay estimates using both methods. The resulting interpretations in dollar terms became more relatable. Metropolitan planning organization can use these results to understand how residents perceive home value in dollar terms in the context of location-based attributes for homes.

The spread of fruiting tree species is strongly determined by the behavior and range of fruit-eating animals, particularly birds. Birds either consume and digest seeds or carry and cache them at some distance from the source tree. These carried and settled seeds provide some form of distribution which generates tree spread to the new location. Firstly, we modal seed dispersal by birds and introduce it in a dispersal model to estimate seed distribution. Using this distribution, we create a population model to estimate the speed at which juniper and pinyon forest boundaries move. Secondly, we introduce a fact that bird movement occurs based on local habitat type to receive modified dispersal model. Birds can easily move many kilometers but habitat changes on the scale of tens of meters with rapidly varying. We develop a new technique to solve the modified dispersal model and approximate the form of transported seed distributions in highly variable landscapes. Using a tree population model, we investigate the rate of forest migration in variable landscapes. We show that speeds calculated using average motility of animals and mean seed handling times accurately predict the migration rate of trees. Regional scale forest distribution models are frequently used to project tree migration based on climate and geographic variables such as elevation, and regional presence-absence data. It is difficult to accurately use dispersal models based on large-scale presence-absence data, particularly for tree species dispersed by birds. The challenge is that variables associated with seed dispersal by birds are represented only few meters while the smallest pixel size for the distribution models begins with few kilometers. Transported seed distribution estimated in the variable landscape offers a tool to make use of this scale separation. Finally, we develop a scenarios that allows us to find large scale dispersal probabilities based on small scale environmental variables.

A method to detect relationships between disease susceptibility and multilocus genetic interactions is the Multifactor-Dimensionality Reduction (MDR) technique pioneered by Ritchie et al. (2001). Since its introduction, many extensions have been pursued to deal with non-binary outcomes and/or account for multiple interactions simultaneously. Studying the effects of multilocus genetic interactions on continuous traits (blood pressure, weight, etc.) is one case that MDR does not handle. Culverhouse et al. (2004) and Gui et al. (2013) proposed two different methods to analyze such a case. In their research, Gui et al. (2013) introduced the Quantitative Multifactor-Dimensionality Reduction (QMDR) that uses the overall average of response variable to classify individuals into risk groups. The classification mechanism may not be efficient under some circumstances, especially when the overall mean is close to some multilocus means. To address such difficulties, we propose a new algorithm, the Ordered Combinatorial Quantitative Multifactor-Dimensionality Reduction (OQMDR), that uses a series of testings, based on ascending order of multilocus means, to identify best interactions of different orders with risk patterns that minimize the prediction error. Ten-fold cross-validation is used to choose from among the resulting models. Regular permutations testings are used to assess the significance of the selected model. The assessment procedure is also modified by utilizing the Generalized Extreme-Value distribution to enhance the efficiency of the evaluation process. We presented results from a simulation study to illustrate the performance of the algorithm. The proposed algorithm is also applied to a genetic data set associated with Alzheimer's Disease.

Software engineering projects that utilize inappropriate pathfinding algorithms carry a significant risk of poor runtime performance for customers. Using social network theory, this experimental study examined the impact of algorithms, frameworks, and map complexity on elapsed time and computer memory consumption. The 1,800 2D map samples utilized were computer random generated and data were collected and processed using Python language scripts. Memory consumption and elapsed time results for each of the 12 experimental treatment groups were compared using factorial MANOVA to determine the impact of the 3 independent variables on elapsed time and computer memory consumption. The MANOVA indicated a significant factor interaction between algorithms, frameworks, and map complexity upon elapsed time and memory consumption, F(4, 3576) = 94.09, p < .001, h2 = .095. The main effects of algorithms, F(4, 3576) = 885.68, p < .001, h2 = .498; and frameworks, F(2, 1787) = 720,360.01, p .001, h2 = .999; and map complexity, F(2, 1787) = 112,736.40, p < .001, h2 = .992, were also all significant. This study may contribute to positive social change by providing software engineers writing software for complex networks, such as analyzing terrorist social networks, with empirical pathfinding algorithm results. This is crucial to enabling selection of appropriately fast, memory-efficient algorithms that help analysts identify and apprehend criminal and terrorist suspects in complex networks before the next attack.

This thesis explores the problems of fitting antedependence (AD) models and partial antecorrelation (PAC) models to continuous non-Gaussian longitudinal data. AD models impose certain conditional independence relations among the measurements within each subject, while PAC models characterize the partial correlation relations. The models are parsimonious and useful for data exhibiting time-dependent correlations. Since the relation of conditional independence among variables is rather restrictive, we first consider an autoregressively characterized PAC model with independent asymmetric Laplace (ALD) innovations and prove that this model is an AD model. The ALD distribution previously has been applied to quantile regression and has shown promise for modeling asymmetrically distributed ecological data. In addition, the double exponential distribution, a special case of the ALD, has played an important role in fitting symmetric finance and hydrology data. We give the distribution of a linear combination of independent standard ALD variables in order to derive marginal distributions for the model. For the model estimation problem, we propose an iterative algorithm for the maximum likelihood estimation. The estimation accuracy is illustrated by some numerical examples as well as some longitudinal data sets. The second component of this dissertation focuses on AD multivariate skew normal models. The multivariate skew normal distribution not only shares some nice properties with multivariate normal distributions but also allows for any value of skewness. We derive necessary and sufficient conditions on the shape and covariance parameters for multivariate skew normal variables to be AD(p) for some p. Likelihood-based estimation for balanced and monotone missing data as well as likelihood ratio hypothesis tests for the order of antedependence and for zero skewness under the models are presented. Since the class of skew normal random variables is closed under the addition of independent standard normal random variables, we then consider an autoregressively characterized PAC model with a combination of independent skew normal and normal innovations. Explicit expressions for the marginals, which all have skew normal distributions, and maximum likelihood estimates of model parameters, are given. Numerical results show that these three proposed models may provide reasonable fits to some continuous non-Gaussian longitudinal data sets. Furthermore, we compare the fits of these models to the Treatment A cattle growth data using penalized likelihood criteria, and demonstrate that the AD(2) multivariate skew normal model fits the data best among those proposed models.

This study retrofitted a Diagnostic Classification Model (DCM) known as the Fusion model onto non-diagnostic test data from of the University of Chicago School Mathematics Project (UCSMP) Algebra and Geometry Readiness test post-test used with Transition Mathematics (Third Edition, Field-Trial Version). The test contained 24 multiple-choice middle school math items, and was originally given to 95 advanced 6th grade and 293 7th grade students. The use of these test answers for this study was an attempt to show that by using cognitive diagnostic analysis techniques on test items not constructed for that purpose, highly predictable multidimensional cognitive attribute profiles for each test taker could be obtained. These profiles delineated whether a given test taker was a master or non-master for each attribute measured by the test, thus allowing detailed diagnostic feedback to be disseminated to both the test takers and their teachers. The full version of the non-compensatory Fusion model, specifically, along with the Arpeggio software package, was used to estimate test taker profiles on each of the four cognitive attributes found to be intrinsic to the items on this test, because it handled both slips and guesses by test takers and accounted for residual skills not defined by the four attributes and twenty-four items in the Q-matrix. The attributes, one or more of which was needed to correctly answer an item, were defined as: Skills— those procedures that students should master with fluency; e.g., multiplying positive and negative numbers; Properties—which deal with the principles underlying the mathematics concepts being studied, such as being able to recognize and use the Repeated-Addition Property of Multiplication; Uses—which deal with applications of mathematics in real situations ranging from routine "word problems" to the development and use of mathematical models, like finding unknowns in real situations involving multiplication; and, Representations—which deal with pictures, graphs, or objects that illustrate concepts. Ultimately, a Q-matrix was developed from the rating of four content experts, with the attributes needed to answer each item clearly delineated. A validation of this Q-matrix was obtained from the Fusion model Arpeggio application to the data as test taker profiles showed which attributes were mastered by each test taker and which weren’t. Masters of the attributes needed to be acquired to successfully answer a test item had a proportion-correct difference from non-masters of .44, on average. Regression analysis produced an R-squared of .89 for the prediction of total scores on the test items by the attribute mastery probabilities obtained from the Fusion model with the final Q-matrix. Limitations of the study are discussed, along with reasons for the significance of the study.