Ingeniero Civil Industrial / Una entidad bancaria nacional propuso el problema de pronosticar probabilidades de incumplimiento crediticio a lo largo del tiempo (PI), motivada por cambios en estándares internacionales y por buscar una visión extendida del riesgo de sus clientes de consumo. Dentro de los objetivos del proyecto destacan: determinar una expresión para la PI como función del tiempo, evaluar y discriminar los mejores modelos de acuerdo a criterios estadísticos y operativos, y crear curvas específicas para diferentes segmentos de clientes. El enfoque utilizado para la predicción fue el análisis de supervivencia, el cual modela de forma probabilística el tiempo hasta que ocurre un incumplimiento. Para ello se contó con un panel de 10.000 clientes, el cual consigna para cada uno de ellos, su tiempo de supervivencia, su estatus de incumplimiento e información longitudinal de variables tales como: la mora, la antigüedad de la cuenta corriente, el score crediticio y el estatus de renegociación. La metodología consistió en tres etapas: pre-procesamiento y transformación, exploración y procesamiento de datos. En la primera se aplicaron los filtros de datos, en la segunda se buscaron patrones en ellos y en la tercera etapa se entrenaron los modelos. Se evaluaron los modelos mediante criterios estadísticos y operativos. En los primeros figuran el criterio de información de Aikake y los residuos de Cox-Snell. Mientras que los segundos se basaron en la adaptación de los modelos a los procesos originación (ingreso de nuevos clientes) y seguimiento de clientes (evaluación a lo largo del tiempo). Los resultados indican que la mora y la renegociación actúan en favor del incumplimiento, mientras que el score y la antigüedad lo hacen en contra. Esto se cumplió para la mayoría de los modelos con un 99% de confianza. Dentro de las conclusiones se destaca que el mejor modelo de originación es la regresión AFT lognormal con covariables evaluadas al inicio, porque la distribución de sus residuos fue la más cercana a una exponencial de tasa 1 y porque evalúa a nuevos clientes de acuerdo a un modelo entrenado con individuos de su misma condición. Mientras que el mejor modelo de seguimiento corresponde a una regresión de Cox con covariables dependientes del tiempo, ya que incorpora la historia de los clientes y fue el modelo con menor AIC dentro de los modelos semi-parametricos. Finalmente se destaca que el incumplimiento crediticio se puede interpretar como un fenómeno de supervivencia, cuyo enfoque también admite extensiones, tales como la incorporación del pago temprano como nuevo evento y la inclusión de variables macroeconómicas. Estos aspectos se propondrán para futuros desarrollos. / Este trabajo ha sido parcialmente financiado por Banco BCI

No description available.

This thesis deals with problematic of thermodynamic stability of AFt phases. It is focused on studium of mineral thaumasite, mainly its possible ways of synthesis. Specific aim of the task is to prepare thaumasite by hydratation of mineral ternesite in optimal condition and consequential evalution by X-ray diffraction, DTA and SEM.

This thesis deals with monitoring the thermodynamic stability of thaumasite and monitoring it is properties in the long term. The aim was to synthesize thaumasite under different conditions and to evaluate it using X-ray diffraction analysis and thermal analysis.

The thesis is focused on the optimization of the thaumasite preparation in the way of ternesite hydration and further comparison of this approach to the other methods. The thaumasite was prepared by three different methods for this purpose - according to Aguilera et al., Purnell and the hydration method of ternesite clinkers. Finally, the X-ray diffraction analysis was particularly used to evaluate the obtained properties and results.

In survival analysis or time-to-event analysis, one of the primary goals of analysis is to predict the occurrence of an event of interest for subjects within the study. Even though survival analysis methods were originally developed and used in medical re- search, those methods are also commonly used nowadays in other areas as well, such as in predicting the default of a loan and in estimating of the failure of a system. To include covariates in the analysis, the most widely used models are the propor- tional hazard model developed by Cox (1972) and the accelerated failure time model developed by Buckley and James (1979). The proportional hazard (PH) model as- sumes subjects from different groups have their hazard functions proportionally, while the accelerated failure time (AFT) model assumes the effect of covariates is to accel- erate or decelerate the occurrence of event of interest. In some survival analyses, not all subjects in the study will experience the event. Such a group of individuals is referred to `cured' group. To analyze a data set with a cured fraction, Boag (1948) and Berkson and Gage (1952) discussed a mixture cure model. Since then, the cure model and associated inferential methods have been widely stud- ied in the literature. It has also been recognized that subjects in the study are often correlated within clusters or groups; for example, patients in a hospital would have similar conditions and environment. For this reason, Vaupel et al. (1979) proposed a frailty model to model the correlation among subjects within clusters and conse- quently the presence of heterogeneity in the data set. Hougaard (1989), McGilchrist and Aisbett (1991), and Klein (1992) all subsequently developed parametric frailty models. Balakrishnan and Peng (2006) proposed a Generalized Gamma frailty model, which includes many common frailty models, and discussed model fitting and model selection based on it. To combine the key components and distinct features of the mixture cure model and the frailty model, a mixture cure frailty model is discussed here for modelling correlated survival data when not all the subjects under study would experience the occurrence of the event of interest. Longini and Halloran (1996) and Price and Manatunga (2001) developed several parametric survival models and employed the Likelihood Ratio Test (LRT) to perform a model discrimination among cure, frailty and mixture cure frailty models. In this thesis, we first describe the components of a mixture cure frailty model, wherein the flexibility of the frailty distributions and lifetime survival functions are discussed. Both proportional hazard and accelerated failure time models are considered for the distribution of lifetimes of susceptible (or non-cured) individuals. Correlated ran- dom effect is modelled by using a Generalized Gamma frailty term, and an EM-like algorithm is developed for the estimation of model parameters. Some Monte Carlo simulation studies and real-life data sets are used to illustrate the models as well as the associated inferential methods. / Thesis / Doctor of Philosophy (PhD)

Up to present, ettrigite has been prepared by two means in the Institute of Technology of Building Materials and Components of the BUT Faculty of Civil Engineering: originally by the hydration of the yeelimite clinker; recently by the direct synthesis of ettringite from calcium hydroxide and aluminum sulfate, the latter being considered more suitable in qualitative and quantitative terms. This diploma thesis focuses on the comparison of both means, using ettringite produced by the latter as control sample. This endeavor focuses on monitoring the stability of ettringite and its possible conversion to other minerals using X-ray diffraction analysis, thermal analysis and scanning electron microscopy (SEM).

The diploma thesis deals with development and properties of historic mortars, evolves and optimizes recipe of historicizing universal mortar with mixture binder, which have been evolves in previous phase. Binder system is based on lime hydrate and metakaolin and is modified with fast-pacing gypsum. Thesis follow up technological properties of mixtures of historicizing mortars and modifies them for current construction industry. It also follows effect of modification to phase composition of binder system and eventually rise of AFt phases.

This diploma thesis is focused on long term monitoring of thermodynamic stability of ettringite under selected conditions and its possible destabilization or transformation into another AFt phase, especially thaumasite. Ettringite samples were made in two ways, by hydrating the yeelimite in the system with the alite and by addition of aluminium sulphate and calcium hydroxide.

The thesis is oriented on monitoring of hydration process of portland cement based on fluidized bed ash, firstly on investigation of AFt phases, mainly ettringite and thaumasite. Specific aim of the task is to prepare the scheme of these minerals synthetic preparation and to verify their laboratory preparation by different ways, using methods of RTG-diffraction analysis and scanning electron microscopy.