Anwendung des EM-Algorithmus zur Schätzung der Parameter von Fluoreszenzlebensdauerfunktionen

Die Bachelorarbeit behandelt die Schätzung der Parameter von Fluoreszenzlebensdauerfunktionen mit Hilfe des EM-Algorithmus. Dabei wird der Algorithmus sowohl auf simulierte als auch auf gemessene Daten angewandt. Die Schätzung der Parameter erfolgt zunächst global für die gesamte Probe mit Hilfe eines Simplex-Verfahrens, um dann das Verhältnis der Komponenten der Fluoreszenzlebensdauer, also die Wahrscheinlichkeit, mit der ein Photon von einer Komponente stammt, für jedes Pixel eines Bildes durch den EM-Algorithmus zu bestimmen. Die Messungen liegen als Anzahl der gemessenen Photonen in diskreten Zeitintervallen vor, dabei fehlt jedoch die Information, wie viele der Photonen in einem der Intervalle zu einer Komponente gehören. Durch die Nutzung bedingter Erwartungswerte ist der EM-Algorithmus in der Lage, ohne Verzerrung mit diesen unbekannten Daten umzugehen. Weiterhin wird die Schätzung dadurch erschwert, dass die Daten durch Faltung der Fluoreszenzlebensdauerfunktion mit einer so genannten Apparatefunktion zustandekommen und das Modell somit sehr komplex wird. Auch für dieses Problem wird im Laufe der Arbeit eine Lösung vorgestellt.

Entwicklung nachhaltiger Synthesemethoden zur atomökonomischen Transformation von 1-Vinylpropargylalkoholen mit Übergangsmetallkomplexen redoxaktiver Liganden als Katalysatoren

Magdeburg, Univ., Fak. für Verfahrens- und Systemtechnik, Diss., 2014

Optimal experimental design for parameter identification and model selection

Magdeburg, Univ., Fak. für Elektrotechnik und Informationstechnik, Diss., 2014

Untersuchungen zur Perzeption musikalischer Parameter von Cochlea-Implantat-Trägern mit den Sprachkodierungsstrategien ACE und MP3000 im Vergleich zu normalhörenden Probanden

Magdeburg, Univ., Med. Fak., Diss., 2014

Aging and growth parameter from the Piaractus mesopotamicus (pacu) from the Cuiabá river, Mato Grosso, Brazil

This study has aims to determine the age and to estimate the growth parameters using scales of the species. Individuals of Piaractus mesopotamicus (Holmberg, 1887) used in this study were captured in the commercial fishery conducted in the region, along the year 2006. The model selected to express the growth of the species was the von Bertalanffy Sl= Sl∞*[1-exp-k(t-to)]. To determine if scales are suitable for studying the growth of pacu, we analyzed the relation between standard length (Sl) and the radius of the scales through linear regression. The period of annuli formation was determined analyzing the variations in the marginal increment and evaluating the consistency of the readings through the analysis of the coefficient of variations (CVs) for the average standard lengths of each age (number of rings) observed in the scales. The relationship between Ls of the fish and the radius of the scales showed that scales can be used to study the age and growth of P. mesopotamicus (R= 0.79). CVs were always below 20%, demonstrating the consistency of the readings. Annuli formation occurred in February, probably related to trophic migration that occurs in this month in the region. Equations that represents the growth in length obtained for P. mesopotamicus are Sl=50.00*[1-exp-0.18(t-(-3.00)] for males and Sl=59.23*[1-exp-0.14(t-(-3.36)] for females. The growth parameters obtained in this study were lower compared to other studies previously conducted for the same species and can related to overexploitation that species is submitted by fishing in the region. These values show also that females of pacu attain greater asymptotic length than males that growth faster.

Malignant transformation of a rat fibroma by the treatment with an anti-fibrosing drug: CY-168F (Plastenan)

Fifteen albino (Sprague Dawley) rats with subcutaneous transplanted fibromas was used in the present study. The tumour was formed by typical fibroblasts in a dense collagen matrix and was provenient from a fibroma that appeared spontaneously in an albino rat of the same strain. Ultrastructurally collagen disclosed normal periodicity and the fibroblasts showed irregular notched nuclei with irregular distribution of chromatin, that suggests transitional aspects to fibrosarcoma. The 15 animals, from different passage groups, were divided into: 8 animals submitted to treatment with the drug acexamic acid (CY-168F) - N acetyl-amino-6-hexanoic acid (plastenan) and 7 untreated control animals. Three of the treated animals showed a malignant transformation to fibrosarcoma. transitional histological features from typical fibroma to highly indifferentiated fibrosarcoma could be detected in come animal subjected to repeated biopsies. Ultrastructural study disclosed nuclear alterations and hyperactive ergastoplasm and collagen containing inclusions into the cytoplasm of fibroblasts. In the group of 7 untreated naimals, no malignant transformation could be detected histologically. Two aspects deserve attention: the malignant potential of a typical fibroma and the apparent effect of an antifibrosing drug in inducing malignization of this tumour.

Large Time-Varying Parameter VARs

In this paper we develop methods for estimation and forecasting in large timevarying parameter vector autoregressive models (TVP-VARs). To overcome computational constraints with likelihood-based estimation of large systems, we rely on Kalman filter estimation with forgetting factors. We also draw on ideas from the dynamic model averaging literature and extend the TVP-VAR so that its dimension can change over time. A final extension lies in the development of a new method for estimating, in a time-varying manner, the parameter(s) of the shrinkage priors commonly-used with large VARs. These extensions are operationalized through the use of forgetting factor methods and are, thus, computationally simple. An empirical application involving forecasting inflation, real output, and interest rates demonstrates the feasibility and usefulness of our approach.

Model Switching and Model Averaging in Time- Varying Parameter Regression Models

This paper investigates the usefulness of switching Gaussian state space models as a tool for implementing dynamic model selecting (DMS) or averaging (DMA) in time-varying parameter regression models. DMS methods allow for model switching, where a different model can be chosen at each point in time. Thus, they allow for the explanatory variables in the time-varying parameter regression model to change over time. DMA will carry out model averaging in a time-varying manner. We compare our exact approach to DMA/DMS to a popular existing procedure which relies on the use of forgetting factor approximations. In an application, we use DMS to select different predictors in an in ation forecasting application. We also compare different ways of implementing DMA/DMS and investigate whether they lead to similar results.

Hierarchical Shrinkage in Time-Varying Parameter Models

In this paper, we forecast EU-area inflation with many predictors using time-varying parameter models. The facts that time-varying parameter models are parameter-rich and the time span of our data is relatively short motivate a desire for shrinkage. In constant coefficient regression models, the Bayesian Lasso is gaining increasing popularity as an effective tool for achieving such shrinkage. In this paper, we develop econometric methods for using the Bayesian Lasso with time-varying parameter models. Our approach allows for the coefficient on each predictor to be: i) time varying, ii) constant over time or iii) shrunk to zero. The econometric methodology decides automatically which category each coefficient belongs in. Our empirical results indicate the benefits of such an approach.

An Introduction to parametric and non-parametric models for bivariate positive insurance claim severity distributions

We present a real data set of claims amounts where costs related to damage are recorded separately from those related to medical expenses. Only claims with positive costs are considered here. Two approaches to density estimation are presented: a classical parametric and a semi-parametric method, based on transformation kernel density estimation. We explore the data set with standard univariate methods. We also propose ways to select the bandwidth and transformation parameters in the univariate case based on Bayesian methods. We indicate how to compare the results of alternative methods both looking at the shape of the overall density domain and exploring the density estimates in the right tail.

Inequality across countries in energy intensities: an analysis of the role of energy transformation and final energy consumption

This paper analyzes the role of the energy transformation index and of final energy consumption per GDP unit in the disparities in energy intensity across countries. In that vein, we use a Theil decomposition approach to analyze global primary energy intensity inequality as well as inequality across different regions of the world and inequality within these regions. The paper first demonstrates the pre-eminence of divergence in final energy consumption per GDP unit in explaining global primary energy intensity inequality and its evolution during the 1971-2006 period. Secondly, it shows the lower (albeit non negligible) impact of the transformation index in global primary energy inequality. Thirdly, the relevance of regions as unit of analysis in studying crosscountry energy intensity inequality and their explanatory factors is highlighted. And finally, how regions around the world differ as to the relevance of the energy transformation index in explaining primary energy intensity inequality.

A parameter-free approach for solving combinatorial optimization problems through biased randomization of efficient heuristics

This paper discusses the use of probabilistic or randomized algorithms for solving combinatorial optimization problems. Our approach employs non-uniform probability distributions to add a biased random behavior to classical heuristics so a large set of alternative good solutions can be quickly obtained in a natural way and without complex conguration processes. This procedure is especially useful in problems where properties such as non-smoothness or non-convexity lead to a highly irregular solution space, for which the traditional optimization methods, both of exact and approximate nature, may fail to reach their full potential. The results obtained are promising enough to suggest that randomizing classical heuristics is a powerful method that can be successfully applied in a variety of cases.