Statistical analysis of survey-based event history data with application to modeling of unemployment duration

Longitudinal surveys are increasingly used to collect event history data on person-specific processes such as transitions between labour market states. Surveybased event history data pose a number of challenges for statistical analysis. These challenges include survey errors due to sampling, non-response, attrition and measurement. This study deals with non-response, attrition and measurement errors in event history data and the bias caused by them in event history analysis. The study also discusses some choices faced by a researcher using longitudinal survey data for event history analysis and demonstrates their effects. These choices include, whether a design-based or a model-based approach is taken, which subset of data to use and, if a design-based approach is taken, which weights to use. The study takes advantage of the possibility to use combined longitudinal survey register data. The Finnish subset of European Community Household Panel (FI ECHP) survey for waves 1–5 were linked at person-level with longitudinal register data. Unemployment spells were used as study variables of interest. Lastly, a simulation study was conducted in order to assess the statistical properties of the Inverse Probability of Censoring Weighting (IPCW) method in a survey data context. The study shows how combined longitudinal survey register data can be used to analyse and compare the non-response and attrition processes, test the missingness mechanism type and estimate the size of bias due to non-response and attrition. In our empirical analysis, initial non-response turned out to be a more important source of bias than attrition. Reported unemployment spells were subject to seam effects, omissions, and, to a lesser extent, overreporting. The use of proxy interviews tended to cause spell omissions. An often-ignored phenomenon classification error in reported spell outcomes, was also found in the data. Neither the Missing At Random (MAR) assumption about non-response and attrition mechanisms, nor the classical assumptions about measurement errors, turned out to be valid. Both measurement errors in spell durations and spell outcomes were found to cause bias in estimates from event history models. Low measurement accuracy affected the estimates of baseline hazard most. The design-based estimates based on data from respondents to all waves of interest and weighted by the last wave weights displayed the largest bias. Using all the available data, including the spells by attriters until the time of attrition, helped to reduce attrition bias. Lastly, the simulation study showed that the IPCW correction to design weights reduces bias due to dependent censoring in design-based Kaplan-Meier and Cox proportional hazard model estimators. The study discusses implications of the results for survey organisations collecting event history data, researchers using surveys for event history analysis, and researchers who develop methods to correct for non-sampling biases in event history data.

Methodology for wind turbine blade geometry optimization

Nowadays, the upwind three bladed horizontal axis wind turbine is the leading player on the market. It has been found to be the best industrial compromise in the range of different turbine constructions. The current wind industry innovation is conducted in the development of individual turbine components. The blade constitutes 20-25% of the overall turbine budget. Its optimal operation in particular local economic and wind conditions is worth investigating. The blade geometry, namely the chord, twist and airfoil type distributions along the span, responds to the output measures of the blade performance. Therefore, the optimal wind blade geometry can improve the overall turbine performance. The objectives of the dissertation are focused on the development of a methodology and specific tool for the investigation of possible existing wind blade geometry adjustments. The novelty of the methodology presented in the thesis is the multiobjective perspective on wind blade geometry optimization, particularly taking simultaneously into account the local wind conditions and the issue of aerodynamic noise emissions. The presented optimization objective approach has not been investigated previously for the implementation in wind blade design. The possibilities to use different theories for the analysis and search procedures are investigated and sufficient arguments derived for the usage of proposed theories. The tool is used for the test optimization of a particular wind turbine blade. The sensitivity analysis shows the dependence of the outputs on the provided inputs, as well as its relative and absolute divergences and instabilities. The pros and cons of the proposed technique are seen from the practical implementation, which is documented in the results, analysis and conclusion sections.

Design of a 1 kW 150 rpm permanent-magnet synchronous generator for stand-alone wind-power applications

A program for calculating low-speed low-power synchronous machine is presented. A permanent-magnet synchronous generator for 1 kW 150 rpm is designed. Optimization of magnet’s and coil’s dimensions was made.

Efficient Optimization Algorithms for Nonlinear Data Analysis

Identification of low-dimensional structures and main sources of variation from multivariate data are fundamental tasks in data analysis. Many methods aimed at these tasks involve solution of an optimization problem. Thus, the objective of this thesis is to develop computationally efficient and theoretically justified methods for solving such problems. Most of the thesis is based on a statistical model, where ridges of the density estimated from the data are considered as relevant features. Finding ridges, that are generalized maxima, necessitates development of advanced optimization methods. An efficient and convergent trust region Newton method for projecting a point onto a ridge of the underlying density is developed for this purpose. The method is utilized in a differential equation-based approach for tracing ridges and computing projection coordinates along them. The density estimation is done nonparametrically by using Gaussian kernels. This allows application of ridge-based methods with only mild assumptions on the underlying structure of the data. The statistical model and the ridge finding methods are adapted to two different applications. The first one is extraction of curvilinear structures from noisy data mixed with background clutter. The second one is a novel nonlinear generalization of principal component analysis (PCA) and its extension to time series data. The methods have a wide range of potential applications, where most of the earlier approaches are inadequate. Examples include identification of faults from seismic data and identification of filaments from cosmological data. Applicability of the nonlinear PCA to climate analysis and reconstruction of periodic patterns from noisy time series data are also demonstrated. Other contributions of the thesis include development of an efficient semidefinite optimization method for embedding graphs into the Euclidean space. The method produces structure-preserving embeddings that maximize interpoint distances. It is primarily developed for dimensionality reduction, but has also potential applications in graph theory and various areas of physics, chemistry and engineering. Asymptotic behaviour of ridges and maxima of Gaussian kernel densities is also investigated when the kernel bandwidth approaches infinity. The results are applied to the nonlinear PCA and to finding significant maxima of such densities, which is a typical problem in visual object tracking.

Stability Analysis in Multicriteria Discrete Portfolio Optimization.

Almost every problem of design, planning and management in the technical and organizational systems has several conflicting goals or interests. Nowadays, multicriteria decision models represent a rapidly developing area of operation research. While solving practical optimization problems, it is necessary to take into account various kinds of uncertainty due to lack of data, inadequacy of mathematical models to real-time processes, calculation errors, etc. In practice, this uncertainty usually leads to undesirable outcomes where the solutions are very sensitive to any changes in the input parameters. An example is the investment managing. Stability analysis of multicriteria discrete optimization problems investigates how the found solutions behave in response to changes in the initial data (input parameters). This thesis is devoted to the stability analysis in the problem of selecting investment project portfolios, which are optimized by considering different types of risk and efficiency of the investment projects. The stability analysis is carried out in two approaches: qualitative and quantitative. The qualitative approach describes the behavior of solutions in conditions with small perturbations in the initial data. The stability of solutions is defined in terms of existence a neighborhood in the initial data space. Any perturbed problem from this neighborhood has stability with respect to the set of efficient solutions of the initial problem. The other approach in the stability analysis studies quantitative measures such as stability radius. This approach gives information about the limits of perturbations in the input parameters, which do not lead to changes in the set of efficient solutions. In present thesis several results were obtained including attainable bounds for the stability radii of Pareto optimal and lexicographically optimal portfolios of the investment problem with Savage's, Wald's criteria and criteria of extreme optimism. In addition, special classes of the problem when the stability radii are expressed by the formulae were indicated. Investigations were completed using different combinations of Chebyshev's, Manhattan and Hölder's metrics, which allowed monitoring input parameters perturbations differently.

Statistical segmentation methods and color variance analysis of retinal images

In this research, the effectiveness of Naive Bayes and Gaussian Mixture Models classifiers on segmenting exudates in retinal images is studied and the results are evaluated with metrics commonly used in medical imaging. Also, a color variation analysis of retinal images is carried out to find how effectively can retinal images be segmented using only the color information of the pixels.

Effect of pectinolitic enzymes on the physical properties of caja-manga (Spondias cytherea Sonn.) pulp

Cajá-manga, also known as golden apple and hog-plum, is an exotic fruit native from Îles de la Société (French Polynesia), which was first introduced in Brazil in 1985. The pulp of ripe fruit was treated with the commercial enzymatic pool and its effect was evaluated in terms of yield, as well as the physical properties viscosity, turbidity and color (L* values). Response surface methodology was used and three levels were adopted for the independent variables temperature (30, 40, and 50 ºC), incubation time (30, 60 and 90 minutes) and enzyme concentration (0.01, 0.05, 0.09 v/v%). A central composite statistical design was used to guide the experimental work. The enzyme treatment highly increased both juice yield (up to 56%) and color (up to 8.6%) and strongly decreased viscosity (up to 57.4%), clarity (up to 77%) and turbidity (up to 85.5%). Incubation time was the most interacting facto, whereas temperature was the least one. Optimization analysis was carried out to reduce enzyme concentration to a minimum by superposing the contour plots of the tested properties, and the recommended ranges of the variables enzyme concentration, process temperature and incubation time were, respectively, 0.042-0.068%, 47.0-49.0 ºC and 82-90 minutes.

A statistical approach to define some tofu processing conditions

The aim of this work was to make tofu from soybean cultivar BRS 267 under different processing conditions in order to evaluate the influence of each treatment on the product quality. A fractional factorial 2(5-1) design was used, in which independent variables (thermal treatment, coagulant concentration, coagulation time, curd cutting, and draining time) were tested at two different levels. The response variables studied were hardness, yield, total solids, and protein content of tofu. Polynomial models were generated for each response. To obtain tofu with desirable characteristics (hardness ~4 N, yield 306 g tofu.100 g-1 soybeans, 12 g proteins.100 g-1 tofu and 22 g solids.100 g-1 tofu), the following processing conditions were selected: heating until boiling plus 10 minutes in water bath, 2% dihydrated CaSO4 w/w, 10 minutes coagulation, curd cutting, and 30 minutes draining time.

Spirulina platensis: process optimization to obtain biomass

Spirulina platensis is a photoautotrophic mesophilic cyanobacterium. Its main sources of nutrients are nitrate, urea, and ammonium salts. Spirulina cultivation requires temperature, light intensity, and nutrient content control. This microalgae has been studied and used commercially due to its therapeutic and antioxidant potential. In addition, several studies have reported its ability to use CO2, its immune activity, and use as an adjuvant nutritive factor in the treatment of obesity. The objective of this study is the production of biomass of S. platensis using different rates of stirring, nitrogen source, amount of micronutrients, and luminosity. A 2(4) experimental design with the following factors: stirring (120 and 140 RPM), amount of nitrogen (1.5 and 2.5 g/L), amount of micronutrients (0,25 and 0,75 mL/L) (11 and 15 W), and luminosity was used. Fermentation was performed in a 500 mL conical flask with 250 mL of culture medium and 10% inoculum in an incubator with controlled stirring and luminosity. Fermentation was monitored using a spectrophotometer (560 nm), and each fermentation lasted 15 days. Of the parameters studied, luminosity is the one with the highest significance, followed by the amount of nitrogen and the interaction between stirring and micronutrients. Maximum production of biomass for 15 days was 2.70 g/L under the following conditions: luminosity15W; stirring, 120 RPM; source of nitrogen, 1.5 g/L; and micronutrients, 0.75 mL/L.

Extraction optimization of antioxidant polysaccharides from leaves of Gynura bicolor (Roxb. & Willd.) DC

Orthogonal design was employed to study the effect of extraction time, temperature and liquid-to-solid ratio on the production of antioxidant polysaccharides from leaves of Gynura bicolor (PLG). Analysis of variance was performed on the data obtained. The most relevant variable was extraction time. A liquid-solid ratio of 30:1 (v/w), a temperature of 80 °C and an extraction time of 3 h were found to be optimal for PLG. The optimal extraction yield of 4.9% was obtained through additional verification test. Hydroxyl radical-scavenging activity, reducing power and ferrous ion chelating ability of PLG were determined. PLG possess concentration-dependent antioxidant potency and IC50 of PLG was 4.67, 0.24 and 4.31 mg/mL for hydroxyl radical-scavenging and ferric ion chelating abilities as well as reducing power, respectively. The results suggest that G. bicolor polysaccharides could be potential source of natural antioxidant and be contributor to the health benefits of G. bicolor.

Production optimization and characterization of immunomodulatory peptides obtained from fermented goat placenta

The goat placental immunomodulatory peptides were produced by fermentation with Aspergillus Niger. The objective of the present study was to investigate the effects of fermentation parameters (carbon source content, pH, and time) on spleen lymphocyte proliferation for the highest immune activity of the fermentation broth using response surface methodology (RSM). According to the data analysis by the Design-Expert® software, the stimulation index value (23.51%), which is the maximum immune activity, was obtained under the following conditions: content of carbon source 1.97 g·L-1, initial pH 5.0, and 74.43 h of fermentation time. Under the optimized fermentation conditions, at a certain concentration range, the fermentation broth produced a significant effect on the proliferation of mouse spleen lymphocytes. Ultrafiltration technique was performed to separate the fermentation broth with different MW (molecular weight). It was found that peptides in the range of <10 KDa were the main bioactivity fractions for the immunomodulatory and antioxidant activities.

Optimization of microwave drying conditions of two banana varieties using response surface methodology

AbstractOptimization of microwave drying conditions of Luvhele and Mabonde banana varieties were studied using response surface methodology. The drying was performed using a central composite rotatable design for two variables: microwave power level (100, 200 and 300 W) and drying time (40, 26, and 12 min.) for Luvhele; (100, 200 and 300 W) and (42, 27, and 12 min) for Mabonde. The colour and texture (hardness) data were analyzed using ANOVA and regression analysis. The fitness of the models obtained was good as the lack of fit for each of the models was not significant. The coefficient of determination R2 of the models was relatively high, hence the models obtained for the responses were adequate and acceptable. Drying conditions of 178.76 W, 12 min. drying time were found optimum for product quality at a desirability of 0.91 for Luvhele; while 127.67 W, 12 min. with a desirability of 0.86 was predicted for Mabonde. The result of this study could be used as a standard for microwave processing of Luvhele and Mabondebanana varieties.

Optimization of the HS-SPME-GC/MS technique for determining volatile compounds in red wines made from Isabel grapes (Vitis labrusca)

Abstract Brazilian wine production is characterized by Vitis labrusca grape varieties, especially the economically important Isabel cultivar, with over 80% of its production destined for table wine production. The objective of this study was to optimize and validate the conditions for extracting volatile compounds from wine with the solid-phase microextraction technique, using the response surface method. Based on the response surface analysis, it can be concluded that the central point values maximize the process of extracting volatile compounds from wine, i.e., an equilibrium time of 15 minutes, an extraction time of 35 minutes, and an extraction temperature of 30 °C. Esters were the most numerous compounds found under these extraction conditions, indicating that wines made from Isabel cultivar grapes are characterized by compounds that confer a fruity aroma; this finding corroborates the scientific literature.

Preparation and Investigation of Bio-coal Mixture Fuels Used in Large Diesel Engines

Coal slurry was of vital interest during the last century due to its potential as an alternative fuel where liquid fuels were necessary. Recently, environmental impacts of the traditional fuels, similarities of bio-coal to that of coal, and huge bio-coal supply has attracted the attention to prepare bio-coal slurries as a new fuel. Rudolf Diesel who invented the diesel engine on 1895 was of the opinion that diesel engines are capable to use different kinds of fuels due to the special design. He tried some kind of vegetable oil to operate on his IC engine. Recently, due to high energy density and more environmentally friendly fuel, researchers believe that bio-coal slurries could act as a new alternative fuel in large diesel engines. Loads of research on different kinds of bio-coal slurry were done by the other researchers worldwide and a lot of progress to boost slurry’s quality were achieved recently. The present study aims to achieve the ideal condition of different factors affecting on the quality of bio-coal slurry. One charcoal sample and two kinds of torrefied wood were used to investigate and compare the reaction of various factors. The results show a great gap between the quality of slurries made of different samples and more researches are necessary to fully understand the impact of the different parameter and improving the quality.

Design and implementation of machine vision system for the mobile assembly robot

The review of intelligent machines shows that the demand for new ways of helping people in perception of the real world is becoming higher and higher every year. This thesis provides information about design and implementation of machine vision for mobile assembly robot. The work has been done as a part of LUT project in Laboratory of Intelligent Machines. The aim of this work is to create a working vision system. The qualitative and quantitative research were done to complete this task. In the first part, the author presents the theoretical background of such things as digital camera work principles, wireless transmission basics, creation of live stream, methods used for pattern recognition. Formulas, dependencies and previous research related to the topic are shown. In the second part, the equipment used for the project is described. There is information about the brands, models, capabilities and also requirements needed for implementation. Although, the author gives a description of LabVIEW software, its add-ons and OpenCV which are used in the project. Furthermore, one can find results in further section of considered thesis. They mainly represented by screenshots from cameras, working station and photos of the system. The key result of this thesis is vision system created for the needs of mobile assembly robot. Therefore, it is possible to see graphically what was done on examples. Future research in this field includes optimization of the pattern recognition algorithm. This will give less response time for recognizing objects. Presented by author system can be used also for further activities which include artificial intelligence usage.