none

In this paper there are presented models for predicting juvenile hypertension using logistic regression and nonparametric discriminant analysis.

Today's economic environment requires for a greater emphasis to be placed on the development of cost-effective solutions to meet military capability based requirements. The Joint Capabilities Integration and Development System (JCIDS) process is designed to identify materiel and non-materiel solutions to fill defense department capability requirements and gaps. Non-materiel solutions include: Doctrine, Organization, Training, Materiel, Leadership and Education, Personnel, Facilities, and Policy (DOTMLPF-P) changes. JCIDS specifies that all non-materiel solutions be analyzed and recommendations be made accordingly following a capability-based assessment (CBA). Guidance for performing CBA analysis provides minimal information on how to predict training effectiveness and as a result training investments are not properly assessed and considered as a viable alternative. Investigations into the ability to predict versus evaluate training performance and to quantify uncertainty in training system design are two identified gaps in the capability of existing training evaluation methods. To address these issues, a Methodology to Predict and Evaluate the Effectiveness of Training (MPEET) has been developed. To address the gap in predictive capability MPEET uses primary elements of learning theory and instructional design to predict the cost-effectiveness of a training program, and recommends training alternatives based on decision-maker preferences for each of the cost and effectiveness criteria. The use of educational and instructional theory involves developing and ensuring human performance requirements will be met after training. Utility theory is used to derive an overall criterion consisting of both cost and effectiveness attributes. MPEET uses this criterion as a key variable in determining how to properly allocate resources to gain maximum training effectiveness. To address the gap in quantifying uncertainty in training performance, probability theory is used within a modeling and simulation environment to create and evaluate previously deterministic variables. Effectiveness and cost variables are assigned probability distributions that reflect the applicable range of uncertainty. MPEET is a systems engineering based decision-making tool. It enhances the instructional design process, which is rooted in the fields of education and psychology, by adding an objective verification step to determine how well instructional strategies are used in the design of a training program to meet the required learning objectives. A C-130J pilot case study is used to demonstrate the application of MPEET and to show the plausibility of the approach. For the case study, metrics are derived to quantify the requirement for knowledge, skills, and attitudes in the C-130J pilot training system design. Instructional strategies were defined specifically for the C-130J training program. Feasible training alternatives were generated and evaluated for cost and effectiveness. Using information collected from decision-maker preferences for cost and effectiveness variables, a new training program is created and comparisons are made to the original. The case study allows tradeoffs to be performed quantitatively between the variable importance weightings and mean value of the probabilistic variables. Overall, it is demonstrated that MPEET provides the capability to assess the cost and effectiveness of training system design and is an enabler to the inclusion of training as an independent non-materiel alternative solution during the CBA process. Although capability gaps in the defense acquisition process motivated the development of MPEET its applicability extends to any training program following the instructional design process where the assumed constraints are not prohibitive.

In this thesis, I sought to determine whether frequency bands in the human electroencephalogram could be used to predict baseball batting performance. Past electroencephalographic (EEG) studies have found that alpha power in the human electroencephalogram predicts subsequent performance. Specifically, Mathewson and colleagues (2012) found that background brain activity, in particular, frontal alpha, had a direct correlation with one’s ability to learn a video game. Here, we decided to see if a similar result would hold true for baseball batting performance. We used a portable electroencephalographic (EEG) data collection system to record EEG data prior to batting practice. Participants sat quietly in a room with the portable EEG unit affixed to their head. Participants then stared in silence at a fixation cross in the center of a computer screen for 30 seconds and then counted backwards from 1000 by 7’s for 30 seconds as a masking task while background EEG was recorded. Player’s were then immediately given live batting practice and with performance judged by three different coaches on four different criteria. The four criteria were: batting mechanics, power, contact, and the batter’s ability to recognize good and bad pitches. Post-hoc, a frequency decomposition was performed on each participant’s EEG data to obtain power in all frequency bands. A correlation analysis of EEG power and batting performance showed that beta power and not alpha power predicted the subsequent performance of the batter. Importantly, a high correlation and significance show that predicting a batter’s performance with a portable EEG system, specifically the MUSE Headband, is highly plausible. / Graduate / 2018-08-09

Using historical data from the past two seasons of the National Hockey League, three different prediction models based on Poisson regression are developed. The aim is to determine whether taking into account the recent form of a team as well as data from how they have previously performed against their opponent can help make better predictions of how many goals they will score against this opponent and thereby calculate the likelihood of each outcome. The three models are evaluated using different measures, for example comparing the odds yielded by the models against the odds of bookmakers. Different ways to account for recent form are discussed. The paper concludes that using recent form and head-to-head data will indeed improve predictions.

Forty-four alfalfa hays from different cuttings, maturities, and locations were fed to sheep in a digestion study. Subsamples of the hays along with corresponding fecal samples were ground and analyzed for dry matter, (DM), organic matter (OM), crude protein (CP), acid detergent fiber (ADF), neutral detergent fiber (NDF), and permanganate lignin. In vivo digestibility (IVDMD) were also determined fro each hay. The hay and fecal samples were scanned with a near infrared reflectance spectrophotometer. Chemical and digestible data for each hay and fecal sample were entered into the computer and separate multiple regression equations developed. Fifteen other alfalfa hays with known chemical and digestible data were used to test the alfalfa equation. Standard errors of analysis (%) and corresponding r2s were: 3.55 and 0.81, 0.23 and 0.99, 2.44 and 0.94, 0.85 and 0.98, 1.33 and 0.96, 1.13 and 0.96, 1.13 and 0.80 for DDM, CP, IVDMD, ADF, NDF, and lignin, respectively. Thirty fecal samples with known chemical data were used to test the fecal equation. Standard errors of analysis and corresponding r2s were: 0.13 and 0.96, 0.41 and 0.93, 0.87 and 0.96, 1.79 and 0.90, 2.10 and 0.91, 1.46 and 0.90 for DM, OM, CP, ADF, NDF and lignin, respectively. Using the spectral information from fecal samples and the chemical composition of alfalfa samples to analyze other alfalfa samples was examined. Twelve other alfalfa hays with known chemical and digestible data were used to test the fecal equation. Standard errors of analysis (%) and corresponding r2s were: 4.05 and 0.01, 1.54 and 0.48, 1.63 and 0.71, 13.16 and 0.55, 1.43 and 0.35, 6.52 and 0.13, 4.30 and 0.63, 2.36 and 0.09, 5.75 and 0.49 for DM, OM, CP, IVDMD, ash, NDF, ADF, lignin, and DDM respectively. Hay fecal samples were sieved to study the utility of using sieving as a procedure to support chemical data in predicting DDM. Use of sieving (alfalfa and fecal) and chemical data (other than IVDMD) increased the precision of predicting DDM (R2 = 0.76), over using IVDMD and ash concentrations to predict DDM (R2 = 0.66).

Knowledge of water and salt movement in soils is necessary for development of a management scheme for controlling the quality of irrigation return flow. A computer model was developed to predict the distribution of water and salts in the root zone under varying initial and boundary conditions. The model consists of water flow and salt flow sub-models. The water flow sub-model considers the numerical approximation of the general water flow equation with modification for water loss by evapotranspiration (and thus root extraction). The salt flow sub-model considers the mass flow of salts, chemical exchange, precipitation or dissolution of CaCO3, and CaSO4, and formation of undissociated Ca and Mg sulphate. The model was tested under laboratory and field conditions by comparing predicted values with experimental measurements. Satisfactory agreement was noted for the water content distribution in almost all the experiments. The model yielded approximately correct values of total salt distribution in the field and one of the column experiments. The agreement between the measured and predicted values for the two other column experiments was poor. The poor agreement seems to result from the irregular dissolution of the applied powdered salts. The distribution of individual ions was not accurately predicted by the model. The disagreement between the predicted and measured values was large at high salt concentration. Complex ion formation, insufficient description of exchange and activity coefficients at high salt concentration are suggested for this lack of agreement. Further development and field testing of the model are needed.

Canopy transpiration is a major factor determining crop evapotranspiration and energy budgets. Unfortunately the development of robust models of canopy transpiration is hindered by a lack of reliable data due to the difficulties of making canopy-scale measurements. However, measurements of canopy water vapor and carbon fluxes via gas exchange techniques are possible in controlled environments. Simultaneous measurements of transpiration, photosynthesis, and canopy temperature were made in wheat and soybean communities. These data were used to calculate chamber aerodynamic and canopy stomata! conductances, and to model the response of canopy transpiration to CO2concentration and vapor pressure deficit. Canopy stomata! conductance was found to decrease diurnally by 20-30% in well-watered crops grown under constant environmental conditions. The magnitude of this diurnal decrease in the canopy stomata! conductance of wheat and soybean decreased with increasing ambient CO2 concentrations. Eight models describing how canopy stomatal conductance responds to environmental changes were incorporated into a canopy transpiration model. The results and methods developed in this study will allow future physiologically-based canopy transpiration models to incorporate these models for predicting the response of transpiration rates in controlled environments.

Researchers have been developing techniques to predict inlight icing in order to determine aircraft behavior under different icing conditions. A key component of the techniques is the mesh generation strategy. Automated meshing facilitates numerical simulation of ice accretion on realistic aircraft configurations by deforming the surface and volume meshes in response to the evolving ice shape. The objective of this research is to validate an ice accretion strategy for wings, using a previously developed meshing strategy. The intent is to investigate the effect of varying numerical parameters, on the predicted ice shape. Using this framework, results are simulated for rime and glaze ice accretions on a rectangular planform wing with a constant GLC-305 airfoil section. The number of time steps is shown to have a significant effect on the ice shape, depending on the icing time and conditions. Decreasing the height smoothing parameters generally improves the ice shape accuracy.

No description available.