An investigation of imputation methods for discrete databases and multi-variate time series

When it comes to information sets in real life, often pieces of the whole set may not be available. This problem can find its origin in various reasons, describing therefore different patterns. In the literature, this problem is known as Missing Data. This issue can be fixed in various ways, from not taking into consideration incomplete observations, to guessing what those values originally were, or just ignoring the fact that some values are missing. The methods used to estimate missing data are called Imputation Methods. The work presented in this thesis has two main goals. The first one is to determine whether any kind of interactions exists between Missing Data, Imputation Methods and Supervised Classification algorithms, when they are applied together. For this first problem we consider a scenario in which the databases used are discrete, understanding discrete as that it is assumed that there is no relation between observations. These datasets underwent processes involving different combina- tions of the three components mentioned. The outcome showed that the missing data pattern strongly influences the outcome produced by a classifier. Also, in some of the cases, the complex imputation techniques investigated in the thesis were able to obtain better results than simple ones. The second goal of this work is to propose a new imputation strategy, but this time we constrain the specifications of the previous problem to a special kind of datasets, the multivariate Time Series. We designed new imputation techniques for this particular domain, and combined them with some of the contrasted strategies tested in the pre- vious chapter of this thesis. The time series also were subjected to processes involving missing data and imputation to finally propose an overall better imputation method. In the final chapter of this work, a real-world example is presented, describing a wa- ter quality prediction problem. The databases that characterized this problem had their own original latent values, which provides a real-world benchmark to test the algorithms developed in this thesis.

El resultado internacional de las nuevas empresas. Una visión relacional asociada a la orientación al mercado de la red

[ES] Este trabajo profundiza en el estudio de los factores que influyen en la competitividad internacional de las nuevas empresas internacionales y, en consecuencia, en su resultado internacional. Aunando las disciplinas del emprendedurismo y del marketing internacional, se trata de remarcar la importancia del conocimiento relacional a través de la influencia de la orientación al mercado de la red en los resultados internacionales logrados por estas empresas en base al efector mediador de las ventajas competitivas. Los resultados obtenidos del contraste de hipótesis, mediante modelos de ecuaciones estructurales y análisis multi-muestra, confirman que la orientación al mercado de la red resulta determinante en la obtención de resultados internacionales superiores por parte de las nuevas empresas. Esta influencia se produce de forma indirecta a partir del efecto mediador de las ventajas competitivas en diferenciación y costes desarrolladas por las mismas. Este estudio extiende la investigación pasada en torno al emprendedurismo internacional, incluyendo nuevas aportaciones propias de la disciplina del marketing respecto a los antecedentes de la competitividad y los resultados de las nuevas empresas internacionales en los mercados exteriores. Además, los resultados obtenidos animan a emprendedores en el contexto internacional a considerar el valor explícito de otros factores distintos al conocimiento experiencial, que la empresa adquiere de forma gradual conforme se incrementa su experiencia en el mercado exterior, para darse cuenta del valor potencial que el conocimiento relacional asociado a la orientación al mercado de la red tiene como antecedente para la consecución de ventajas competitivas en el mercado internacional.

Data analysis advances in marine science for fisheries management : supervised classification application

169 p. : il. col.

Association between CASP8-652 6N Del Polymorphism (rs3834129) and Colorectal Cancer Risk: Results from a Multi-Centric Study

The common 2652 6N del variant in the CASP8 promoter (rs3834129) has been described as a putative low-penetrance risk factor for different cancer types. In particular, some studies suggested that the deleted allele (del) was inversely associated with CRC risk while other analyses failed to confirm this. Hence, to better understand the role of this variant in the risk of developing CRC, we performed a multi-centric case-control study. In the study, the variant 2652 6N del was genotyped in a total of 6,733 CRC cases and 7,576 controls recruited by six different centers located in Spain, Italy, USA, England, Czech Republic and the Netherlands collaborating to the international consortium COGENT (COlorectal cancer GENeTics). Our analysis indicated that rs3834129 was not associated with CRC risk in the full data set. However, the del allele was under-represented in one set of cases with a family history of CRC (per allele model OR = 0.79, 95% CI = 0.69-0.90) suggesting this allele might be a protective factor versus familial CRC. Since this multi-centric case-control study was performed on a very large sample size, it provided robust clarification of the effect of rs3834129 on the risk of developing CRC in Caucasians.

Weighting social preferences in participatory multi-criteria evaluations: a case study on sustainable natural resource management

28 p.

Cloning and analysis of laccases from acinetobacter baumannii isolates

Laccases (benzenediol : oxygen oxi doreductases; EC 1.10.3.2) are wide spread i n nature. They are usually found in higher plants and fungi (Thurston 19 94; Mayer and Staples 2002), but recently some bacterial laccases have also been found . The first laccase studied was from Rhus vernicifera in 1883, a Japanese lacquer tree, fr om which the name laccase was derived (Yoshida , 1883). These enzymes belong to the group of bl ue multi - copper oxidases (MCOs) . They usually contain four copper atoms located in three distinct sites. Each site reacts differently to light. The Type 1 (T1) site copper atom absorbs intensely at 600 nm and emits the blue light , the Type 2 (T2) site copper atom is not visible in the absorption spectr um and last, the Type 3 (T3) site has two c opper atoms and absorbs at 330 nm ( Santhanam et al . , 2011; Quintanar et al . , 2007 ) . The protei n structure acts as a complex ligand for the catalytic coppers, providing them the right structure where changes between the reduction states are thermodynamically possible (Dub é , 2008 ) . These enzymes oxidize a surprisingly wide variety of organic and inorganic compounds like, diphenols, polyphenols, substituted phenols, diamines and a romatic amines, with concomitant reduction of molecular oxygen to water (Thurston , 1

Unveiling the factors shaping the distribution of widely distributed alpine vertebrates, using multi-scale ecological niche modelling of the bat Plecotus macrobullaris

Several alpine vertebrates share a distribution pattern that extends across the South-western Palearctic but is limited to the main mountain massifs. Although they are usually regarded as cold-adapted species, the range of many alpine vertebrates also includes relatively warm areas, suggesting that factors beyond climatic conditions may be driving their distribution. In this work we first recognize the species belonging to the mentioned biogeographic group and, based on the environmental niche analysis of Plecotus macrobullaris, we identify and characterize the environmental factors constraining their ranges. Distribution overlap analysis of 504 European vertebrates was done using the Sorensen Similarity Index, and we identified four birds and one mammal that share the distribution with P. macrobullaris. We generated 135 environmental niche models including different variable combinations and regularization values for P. macrobullaris at two different scales and resolutions. After selecting the best models, we observed that topographic variables outperformed climatic predictors, and the abruptness of the landscape showed better predictive ability than elevation. The best explanatory climatic variable was mean summer temperature, which showed that P. macrobullaris is able to cope with mean temperature ranges spanning up to 16 degrees C. The models showed that the distribution of P. macrobullaris is mainly shaped by topographic factors that provide rock-abundant and open-space habitats rather than climatic determinants, and that the species is not a cold-adapted, but rather a cold-tolerant eurithermic organism. P. macrobullaris shares its distribution pattern as well as several ecological features with five other alpine vertebrates, suggesting that the conclusions obtained from this study might be extensible to them. We concluded that rock-dwelling and open-space foraging vertebrates with broad temperature tolerance are the best candidates to show wide alpine distribution in the Western Palearctic.

Association between CASP8-652 6N Del Polymorphism (rs3834129) and Colorectal Cancer Risk: Results from a Multi-Centric Study

The common 2652 6N del variant in the CASP8 promoter (rs3834129) has been described as a putative low-penetrance risk factor for different cancer types. In particular, some studies suggested that the deleted allele (del) was inversely associated with CRC risk while other analyses failed to confirm this. Hence, to better understand the role of this variant in the risk of developing CRC, we performed a multi-centric case-control study. In the study, the variant 2652 6N del was genotyped in a total of 6,733 CRC cases and 7,576 controls recruited by six different centers located in Spain, Italy, USA, England, Czech Republic and the Netherlands collaborating to the international consortium COGENT (COlorectal cancer GENeTics). Our analysis indicated that rs3834129 was not associated with CRC risk in the full data set. However, the del allele was under-represented in one set of cases with a family history of CRC (per allele model OR = 0.79, 95% CI = 0.69-0.90) suggesting this allele might be a protective factor versus familial CRC. Since this multi-centric case-control study was performed on a very large sample size, it provided robust clarification of the effect of rs3834129 on the risk of developing CRC in Caucasians.

Analysis of multivariate stochastic signals sampled by on-line particle analyzers: Application to the quantitative assessment of occupational exposure to NOAA in multisource industrial scenarios (MSIS)

In multisource industrial scenarios (MSIS) coexist NOAA generating activities with other productive sources of airborne particles, such as parallel processes of manufacturing or electrical and diesel machinery. A distinctive characteristic of MSIS is the spatially complex distribution of aerosol sources, as well as their potential differences in dynamics, due to the feasibility of multi-task configuration at a given time. Thus, the background signal is expected to challenge the aerosol analyzers at a probably wide range of concentrations and size distributions, depending of the multisource configuration at a given time. Monitoring and prediction by using statistical analysis of time series captured by on-line particle analyzers in industrial scenarios, have been proven to be feasible in predicting PNC evolution provided a given quality of net signals (difference between signal at source and background). However the analysis and modelling of non-consistent time series, influenced by low levels of SNR (Signal-Noise Ratio) could build a misleading basis for decision making. In this context, this work explores the use of stochastic models based on ARIMA methodology to monitor and predict exposure values (PNC). The study was carried out in a MSIS where an case study focused on the manufacture of perforated tablets of nano-TiO2 by cold pressing was performed