No description available.

In a Bayesian perspective, the probabilistic dependencies between hypotheses under consideration and diagnostic pieces of evidence are the only relevant information for probabilistic updating. We investigated whether human probability judgments conform to this assumption, by manipulating the sensory systems involved in the acquisition and processing of information concerning evidence and hypotheses. Hence, we ran five (computer-based) experiments using a variant of the classic book bag and poker chip task (e.g., Phillips & Edwards, 1966). Participants were first presented with pairs of urns A and B filled with a different proportion of balls that turned either red or green in the visual condition, balls that emitted either a low- or high-pitched sound in the auditory condition, and balls that both turned a color and emitted a sound in various cross-modal (i.e., audio-visual) conditions. One urn was then selected at random, some balls were randomly drawn from it, and their color and/or sound were disclosed. Participants’ task was to estimate the probability that each of the two urns has been selected, given the information provided. In Experiments 1 and 2, we compared the probability judgments referring to probabilistically identical visual and auditory scenarios that only differed with regards to the sensory system involved, without finding any difference between the answers provided in the two conditions. In Experiment 3, 4, and 5, the addition of cross-modal scenarios allowed us to investigate the effects on probabilistic updating of synergic (i.e., both visual and auditory evidence individually supported the hypothesis they jointly supported) or contrasting (i.e., either visual and/or auditory evidence individually supported the hypothesis opposite the one they jointly supported) audio-visual evidence. Our results provide evidence in favor of a synergy-contrasting effect, as probability judgments were more accurate in synergic conditions than in contrasting conditions. This suggests that, when perceptual information is acquired through a singular sensory system, probability judgments conform to the Bayesian assumption that the sensory system involved does not play a role in the updating process, whereas the simultaneous presentation of cross-modal information can influence participants’ performance.

Thesis (M.S. in Operations Research)--Naval Postgraduate School, December 2009. / Thesis Advisor: Kress, Moshe. Second Reader: Johnson, Rachel. "December 2009." Description based on title screen as viewed on February 1, 2010. Author(s) subject terms: Bayesian updating model, Atmospheric Threat and Dispersion model, estimation of location and time of a chemical attack, sensor placement. Includes bibliographical references (p. 99). Also available in print.

On August 15, 2007, Pisco, Peru was hit by an earthquake of Magnitude (Mw) = 8.0 which triggered multiple liquefaction induced lateral spreads. The subduction earthquake lasted for approximately 100 seconds and showed a complex rupture. From the geotechnical perspective, the Pisco earthquake was significant for the amount of soil liquefaction observed. A massive liquefaction induced seaward displacement of a marine terrace was observed in the Canchamana complex. Later analysis using the pre- and post-earthquake images showed that the lateral displacements were concentrated only on some regions. Despite the lateral homogeneity of the marine terrace, some cross-sections showed large displacements while others had minimal displacements. The detailed documentation of this case-history makes it an ideal case-study for the determination of the undrained strength of the liquefied soils; hence, the main objective of this research is to use the extensive data from the Canchamana Slide to estimate the shear strength of the liquefied soils. In engineering practice, the undrained strength of liquefied soil is typically estimated by correlating SPT-N values to: 1) absolute value of residual strength, or 2) residual strength ratio. Our research aims to contribute an important data point that will add to the current understanding of the residual strength of liquefied soils. / Master of Science

One goal of every business is to save money, and building strategies that work to minimize spending and maximize profit is key to the success of a company. Cidewalk is a mobile advertisement company that wished to implement an optimal bidding strategy to help reduce the company's cost and in turn maximize their profits. To accomplish this goal we collected market data which was then analyzed to identify the distribution of second best bids, which is the price Cidewalk pays for an advertisement space by bidding in a Vickrey auction. The optimal bidding model was then implemented through a simulation together with Bayesian updating methods to ensure the model would be responsive to changes in the market. The model's performance was evaluated through the simulation and it was discovered that throughout one day the implemented model yielded 17% savings when compared to Cidewalk's current bidding model.

One goal of every business is to save money, and building strategies that work to minimize spending and maximize profit is key to the success of a company. Cidewalk is a mobile advertisement company that wished to implement an optimal bidding strategy to help reduce the company's cost and in turn maximize their profits. To accomplish this goal we collected market data which was then analyzed to identify the distribution of second best bids, which is the price Cidewalk pays for an advertisement space by bidding in a Vickrey auction. The optimal bidding model was then implemented through a simulation together with Bayesian updating methods to ensure the model would be responsive to changes in the market. The model's performance was evaluated through the simulation and it was discovered that throughout one day the implemented model yielded 17% savings when compared to Cidewalk's current bidding model.

Although Glass Fiber Reinforced Polymer (GFRP) has many potential advantages as reinforcement in concrete structures, the loss in tensile strength of the GFRP reinforcing bar can be significant when exposed to the high alkali environments. Much effort was made to estimate the durability performance of GFRP in concrete; however, it is widely believed the data from accelerated aging tests is not appropriate to predict the longer-term performance of GFRP reinforcing bars. The lack of validated long-term data is the major obstacle for broad application of GFRP reinforcement in civil engineering practices. The main purpose of this study is to evaluate the longer-term deterioration rate of GFRP bars embedded in concrete, and to develop an accurate model that can provide better information to predict the longer-term performance of GFRP bars. In previous studies performed by Trejo, three GFRP bar types (V1, V2, and P type) with two different diameters (16 and 19 mm [0.625, and 0.7 in. referred as #5 and #6, respectively]) provided by different manufacturers were embedded in concrete beams. After pre-cracking by bending tests, specimens were stored outdoors at the Riverside Campus of Texas A&M University in College Station, Texas. After 7 years of outdoor exposure, the GFRP bars were extracted from the concrete beams and tension tests were performed to estimate the residual tensile strength. Several physical tests were also performed to assess the potential changes in the material. It was found that the tensile capacity of the GFRP bars embedded in concrete decreased; however, no significant changes in modulus of elasticity (MOE) were observed. Using this data and limited data from the literature, a probabilistic capacity model was developed using Bayesian updating. The developed probabilistic capacity model appropriately accounts for statistical uncertainties, considering the influence of the missing variables and remaining error due to the inexact model form. In this study, the reduction in tensile strength of GFRP reinforcement embedded in concrete is a function of the diffusion rate of the resin matrix, bar diameter, and time. The probabilistic model predicts that smaller GFRP bars exhibit faster degradation in the tensile capacity than the larger GFRP bars. For the GFRP bars, the model indicates that the probability that the environmental reduction factor required by The American Concrete Institute (ACI) and the American Association of State Highway Transportation Officials (AASHTO) for the design of concrete structures containing GFRP reinforcement is below the required value is 0.4, 0.25, and 0.2 after 100 years for #3, #5, and #6, respectively. The ACI 440 and AASHTO design strength for smaller bars is likely not safe.

This study demonstrates a fully Bayesian approach to multilevel/hierarchical linear growth modeling using freely available software. Further, the study incorporates informative prior distributions for fixed effect estimates using an objective approach. The objective approach uses previous sample results to form prior distributions included in subsequent samples analyses, a process referred to as Bayesian updating. Further, a method for model checking is outlined based on fit indices including information criteria (i.e., Akaike information criterion, Bayesian information criterion, and deviance information criterion) and approximate Bayes factor calculations. For this demonstration, five distinct samples of schools in the process of implementing School-Wide Positive Behavior Interventions and Supports (SWPBIS) collected from 2008 to 2013 were used with the unit of analysis being the school. First, the within-year SWPBIS fidelity growth was modeled as a function of time measured in months from initial measurement occasion. Uninformative priors were used to estimate growth parameters for the 2008-09 sample, and both uninformative and informative priors based on previous years' samples were used to model data from the 2009-10, 2010-11, 2011-12, 2012-13 samples. Bayesian estimates were also compared to maximum likelihood (ML) estimates, and reliability information is provided. Second, an additional three examples demonstrated how to include predictors into the growth model with demonstrations for: (a) the inclusion of one school-level predictor (years implementing) of SWPBIS fidelity growth, (b) several school-level predictors (relative socio-economic status, size, and geographic location), and (c) school and district predictors (sustainability factors hypothesized to be related to implementation processes) in a three-level growth model. Interestingly, Bayesian models estimated with informative prior distributions in all cases resulted in more optimal fit indices than models estimated with uninformative prior distributions.

The first essay developed a respondent model of Bayesian updating for a double-bound dichotomous choice (DB-DC) contingent valuation methodology. I demonstrated by way of data simulations that current DB-DC identifications of true willingness-to-pay (WTP) may often fail given this respondent Bayesian updating context. Further simulations demonstrated that a simple extension of current DB-DC identifications derived explicitly from the Bayesian updating behavioral model can correct for much of the WTP bias. Additional results provided caution to viewing respondents as acting strategically toward the second bid. Finally, an empirical application confirmed the simulation outcomes. The second essay applied a hedonic property value model to a unique water quality (WQ) dataset for a year-round, urban, and coastal housing market in South Florida, and found evidence that various WQ measures affect waterfront housing prices in this setting. However, the results indicated that this relationship is not consistent across any of the six particular WQ variables used, and is furthermore dependent upon the specific descriptive statistic employed to represent the WQ measure in the empirical analysis. These results continue to underscore the need to better understand both the WQ measure and its statistical form homebuyers use in making their purchase decision. The third essay addressed a limitation to existing hurricane evacuation modeling aspects by developing a dynamic model of hurricane evacuation behavior. A household’s evacuation decision was framed as an optimal stopping problem where every potential evacuation time period prior to the actual hurricane landfall, the household’s optimal choice is to either evacuate, or to wait one more time period for a revised hurricane forecast. A hypothetical two-period model of evacuation and a realistic multi-period model of evacuation that incorporates actual forecast and evacuation cost data for my designated Gulf of Mexico region were developed for the dynamic analysis. Results from the multi-period model were calibrated with existing evacuation timing data from a number of hurricanes. Given the calibrated dynamic framework, a number of policy questions that plausibly affect the timing of household evacuations were analyzed, and a deeper understanding of existing empirical outcomes in regard to the timing of the evacuation decision was achieved.

No description available.