"This project utilized generalized estimating equations and general linear modeling to model learning curves for sonographer performance in emergency ultrasonography. Performance was measured in two ways: image quality (interpretable vs. possible hindrance in interpretation) and agreement of findings between the sonographer and an expert reviewing sonographer. Records from 109 sonographers were split into two data sets-- training (n=50) and testing (n=59)--to conduct exploratory analysis and fit the final models for analysis, respectively. We determined that the number of scans of a particular exam type required for a sonographer to obtain quality images on that exam type with a predicted probability of 0.9 is highly dependent upon the person conducting the review, the indication of the scan (educational or medical), and the outcome of the scan (whether there is a pathology positive finding). Constructing family-wise 95% confidence intervals for each exam type demonstrated a large amount of variation for the number of scans required both between exam types and within exam types. It was determined that a sonographer's experience with a particular exam type is not a significant predictor of future agreement on that exam type and thus no estimates were made based on the agreement learning curves. In addition, we concluded based on a type III analysis that when already considering exam type related experience, the consideration of experience on other exam types does not significantly impact the learning curve for quality. However, the learning curve for agreement is significantly impacted by the additional consideration of experience on other exam types."

Made available in DSpace on 2016-06-02T20:06:09Z (GMT). No. of bitstreams: 1 5914.pdf: 1420398 bytes, checksum: e5c2a5d7845b5b17b94959ce849fd613 (MD5) Previous issue date: 2014-04-01 / Financiadora de Estudos e Projetos / The real estate market has a key role in the country and counties economy attracting several studies and researches that explains and interpret the numerous transactions performed, and especially to find appropriate ways to define the monetary value. Usually the real estate data modeling is performed through regression models, especially the linear and also the generalized linear models ( Nelder andWedderburn, 1972). Because these data has different characteristics such as heteroscedasticity, non-normality and heterogeneity, the use of these models can suffer limitations, so it is appropriate to use more and more complex models, such as generalized additive models for location, scale and shape GAMLSS (proposed by Rigby & Stasinopoulos (2005), that allows all parameters of the response variable are modeled parametric or non parametric form. In this context and based on a dataset of urban land of São Carlos city in 2005 was estimated the empirical function the value of the land addressing the class of linear models, generalized linear models and the GAMLSS. Alternatively, considering the existence of two types of real estate prices: already sold (observed) and announced (censored), was proposed to the data, using the survival analysis considering censored left and the GAMLSS in the parameter estimation process. A simulation study and a study of local influence was also performed. / O mercado imobiliário possui um papel fundamental na economia do país e municípios atraindo diversos estudos e pesquisas que buscam explicar e interpretar as inúmeras transações realizadas, e principalmente, encontrar maneiras adequadas de determinar seu valor monetário. Geralmente a modelagem de dados imobiliários e feita por meio de modelos de regressão, especialmente os lineares e também, os modelos lineares generalizados (Nelder e Wedder-burn,1972). Por se tratarem de dados com diferentes características, como heterocedasticidade, não normalidade e heterogeneidade, o uso desses modelos podem sofrer limitações, por isso torna-se adequada a utilização de modelos cada vez mais complexos, como por exemplo, os modelos aditivos generalizados para posição, escala e forma (GAMLSS) propostos por Rigby & Stasinopoulos (2005), que permitem que todas as estimativas dos parâmetros envolvidos no modelo sejam obtidas de forma paramétrica ou não-paramétrica. Neste contexto e com base em um conjunto de dados de lotes urbanos da cidade de Sao Carlos do ano de 2005 foi estimado a função empírica do valor de lotes abordando a classe de modelos lineares, modelos lineares generalizados e o GAMLSS. Alternativamente, considerando a existência de dois tipos de preços de imóveis: ja vendidos (observados) e anunciados (censurados), foi proposto aos dados, a utilização da analise de sobrevivência considerando censura a esquerda e o GAMLSS no processo de estimação dos parâmetros. Foi realizado também um estudo de simulação e um estudo de influência local.