This dissertation is composed of three research papers that deal with the topic of “accounting and innovation” and provide evidence for the area of innovation from two accounting perspectives, i.e., the external disclosure perspective and the internal management control system perspective. The disclosure perspective is addressed in the first paper. Using content analysis, it examines the innovation capital disclosure characteristics, i.e., disclosure quantity and quality, in intellectual capital statements of 51 European for-profit firms. Additionally, the relationship between innovation capital disclosure characteristics and industry, firm size, region of domicile, and disclosure guidelines adopted are analyzed. The second and third paper address the management control system perspective. The second paper contributes to the research on management control systems in product development by exploring the performance effects and the interplay of the levers of control, i.e., interactive, diagnostic, beliefs, and boundary control systems, based on data from a survey of 468 senior managers from the manufacturing industry. Therefore, I compare by structural equation modeling a base model of unrelated levers of control (additive model) with (a) a model of mutual association of levers of control (interdependence model), (b) a model of joint use of levers of control (interaction model), and (c) a combined model of mutual association and joint use of levers of control (combined interdependence/interaction model). The best fitting and most parsimonious model is analyzed in terms of the performance effects of the levers of control. Based on the sample of the second paper, the third paper contributes to the recent literature on the relationship between management control systems and innovation by considering how the four levers of control are used as “packages” in product development. I employ cluster analysis to determine how the levers of control are combined, depending on the type of strategy formation (i.e., intended or emergent) and the degree of innovativeness of the firm. Furthermore, I explore which of these combinations are equifinal in terms of product development and organizational performance.

<p>Klaster analiza ima dugu istoriju i mada se<br />primenjuje u mnogim oblastima i dalje ostaju<br />značajni izazovi. U disertaciji je prikazan uvod<br />u neglatki optimizacioni pristup u<br />klasterovanju, sa osvrtom na problem<br />klasterovanja velikih skupova podataka.<br />Međutim, ovi optimizacioni algoritmi bolje<br />funkcioni&scaron;u u radu sa neprekidnim podacima.<br />Jedan od glavnih izazova u klaster analizi je<br />rad sa velikim skupovima podataka sa<br />kategorijalnim i kombinovanim (numerički i<br />kategorijalni) tipovima promenljivih. Rad sa<br />velikim brojem instanci (objekata) i velikim<br />brojem dimenzija (promenljivih), može<br />predstavljati problem u klaster analizi, zbog<br />vremenske složenosti. Jedan od načina<br />re&scaron;avanja ovog problema je redukovanje broja<br />instanci, bez gubitka informacija.<br />Prvi cilj disertacije je bio upoređivanje<br />rezultata klasterovanja na celom skupu i<br />prostim slučajnim uzorcima sa kategorijalnim i<br />kombinovanim podacima, za različite veličine<br />uzorka i različit broj klastera. Nije utvrđena<br />značajna razlika (p&gt;0.05) u rezultatima<br />klasterovanja na uzorcima obima<br />0.03m,0.05m,0.1m,0.3m (gde je m obim<br />posmatranog skupa) i celom skupu.<br />Drugi cilj disertacije je bio konstrukcija<br />efikasnog postupka klasterovanja velikih<br />skupova podataka sa kategorijalnim i<br />kombinovanim tipovima promenljivih.<br />Predloženi postupak se sastoji iz sledećih<br />koraka: 1. klasterovanje na prostim slučajnim<br />uzorcima određene kardinalnosti; 2.<br />određivanje najboljeg klasterskog re&scaron;enja na<br />uzorku, primenom odgovarajućeg kriterijuma<br />validnosti; 3. dobijeni centri klastera iz ovog<br />uzorka služe za klasterovanje ostatka skupa.<br />Treći cilj disertacije predstavlja primenu<br />klaster analize u definisanju klastera<br />bihejvioralnih faktora rizika u populaciji<br />odraslog stanovni&scaron;tva Srbije, kao i analizu<br />sociodemografskih karakteristika dobijenih<br />klastera. Klaster analiza je primenjena na<br />velikom reprezentativnom uzorku odraslog<br />stanovni&scaron;tva Srbije, starosti 20 i vi&scaron;e godina.<br />Izdvojeno je pet jasno odvojenih klastera sa<br />karakterističnim kombinacijama bihejvioralnih<br />faktora rizika: Bez rizičnih faktora, &Scaron;tetna<br />upotreba alkohola i druge rizične navike,<br />Nepravilna ishrana i druge rizične navike,<br />Nedovoljna fizička aktivnost, Pu&scaron;enje. Rezultati<br />multinomnog logističkog regresionog modela<br />ukazuju da ispitanici koji nisu u braku, lo&scaron;ijeg<br />su materijalnog stanja, nižeg obrazovanja i žive<br />u Vojvodini imaju veću &scaron;ansu za prisustvo<br />vi&scaron;estrukih bihejvioralnih faktora rizika.</p> / <p>The cluster analysis has a long history and a<br />large number of clustering techniques have<br />been developed in many areas, however,<br />significant challenges still remain. In this<br />thesis we have provided a introduction to<br />nonsmooth optimization approach to clustering<br />with reference to clustering large datasets.<br />Nevertheless, these optimization clustering<br />algorithms work much better when a dataset<br />contains only vectors with continuous features.<br />One of the main challenges is clustering of large<br />datasets with categorical and mixed (numerical<br />and categorical) data. Clustering deals with a<br />large number of instances (objects) and a large<br />number of dimensions (variables) can be<br />problematic because of time complexity. One of<br />the ways to solve this problem is by reducing<br />the number of instances, without the loss of<br />information.<br />The first aim of this thesis was to compare<br />the results of cluster algorithms on the whole<br />dataset and on simple random samples with<br />categorical and mixed data, in terms of validity,<br />for different number of clusters and for<br />different sample sizes. There were no<br />significant differences (p&gt;0.05) between the<br />obtained results on the samples of the size of<br />0.03m,0.05m,0.1m,0.3m (where m is the size of<br />the dataset) and the whole dataset.<br />The second aim of this thesis was to<br />develop an efficient clustering procedure for<br />large datasets with categorical and mixed<br />(numeric and categorical) values. The proposed<br />procedure consists of the following steps: 1.<br />clustering on simple random samples of a given<br />cardinality; 2. finding the best cluster solution<br />on a sample (by appropriate validity measure);<br />3. using cluster centers from this sample for<br />clustering of the remaining data.<br />The third aim of this thesis was to<br />examine clustering of four lifestyle risk factors<br />and to examine the variation across different<br />socio-demographic groups in a Serbian adult<br />population. Cluster analysis was carried out on<br />a large representative sample of Serbian adults<br />aged 20 and over. We identified five<br />homogenous health behaviour clusters with<br />specific combination of risk factors: &#39;No Risk<br />Behaviours&#39;, &#39;Drinkers with Risk Behaviours&#39;,<br />&#39;Unhealthy diet with Risk Behaviours&#39;,<br />&#39;Smoking&#39;. Results of multinomial logistic<br />regression indicated that single adults, less<br />educated, with low socio-economic status and<br />living in the region of Vojvodina are most likely<br />to be a part of the clusters with a high-risk<br />profile.</p>

This thesis deals with differences between individual states of the United States from the demographic point of view, and searches for causes of these differences. United States are composed of several disparate areas, which are different from each other in their location, size, number and composition of the population, and historical evolution. All this affects the demographic characteristics of those territorial units. Probably the main cause of differentiation of individual states is racial composition, since the intensity of demographic events is specific to each race or ethnicity. The states differ in levels of fertility, mortality, marriage or education. To confirm these assumptions, the statistical method canonical correlation was used. Using cluster analysis has revealed that there are groups of states that have similar demographic profile. Most notably, it shows the influence of ethnic and racial composition in the south of the United States, where the highest proportion of populations is composed by black race and hispanic origin. The differentiation of levels of infant mortality and life expectancy are mainly influenced. Key words: United States of America, race, ethnicity, black population, hispanic population, total fertility rate, life expectancy, canonical correlation, cluster analysis

Sportanalys definieras av Alamar (2013) som användning av historisk data för att applicera modeller som kan ge information till beslutstagare inom en viss organisation. Det ger dem möjlighet att assistera sin organisation för att få en sportslig fördel. I den här studien utförs sportanalys, mer specifikt analyseras hörnsekvenser inom svensk elitfotboll. En hörnsekvens är den sekvens av händelser som sker från att bollen sätts i spel från hörnans startposition tills det att något av följande villkor uppfylls: 8 händelser sker givet att hörnan slås kort 6 händelser sker givet att hörnan slås långt 15 sekunder passerar Försvarande lag tar över bollen Något lag utför ett regelbrott Skott utförs av attackerande lag Datamaterialet som används är framtaget av företaget Wyscout och tillhandahållet av Football Analytics Sweden AB. De ligor och säsonger som betraktas är de svenska herrligorna Allsvenskan och Superettan för säsongerna 2017, 2018 och 2019. I datamaterialet erhålls information om varje händelse som sker under matchen. Utifrån information om händelsen samt koordinater om var händelsen sker framställs variabler som ska kunna beskriva vad som sker inom en hörnsekvens. Syftet med studien är att först identifiera hörnsekvenser med liknande egenskaper och gruppera dem. Utifrån gruppindelningen undersöks sedan sannolikheten för att en hörnsekvens leder till skott samt vad som påverkar sannolikheten. Algoritmen Partitioning Around Medoids (PAM) används med avståndsmåttet Gower och utvärderingsmåttet silhouette för att identifiera följande fem hörnsekvenstyper: Utåtskruvade hörnor från vänsterhörn med tendens mot främre stolpen och relativt nära mållinjen. Inåtskruvade hörnor från vänsterhörn med tendens mot främre stolpen och längre förflyttning ut från mållinjen. Utåtskruvade hörnor från högerhörn med tendens mot främre stolpen och relativt nära mållinjen. Korta varianter som har längre varaktighet, innefattar fler händelser och involverar fler spelare. Inåtskruvade hörnor från högerhörn med tendens mot främre stolpen och längre förflyttning ut från mållinjen. Betraktas förekomsten av skott i datamaterialet givet klusterstrukturen konstateras att hörnsekvenstyp 4 i störst utsträckning lett till skott med förekomsten 19 procent inom klustret. Hörnsekvenstyperna 2 och 5 är något sämre med respektive 18 procent av hörnsekvenserna som lett till skott. Med dessa hörnsekvenstyper i fokus anpassas flera Bayesianska hierarkiska logitmodeller för att undersöka sannolikheten för att en hörnsekvens leder till skott givet de framtagna variablerna. Vid skapandet av modellerna undersöktes om en hierarkisk modellstruktur var behövlig för att undersöka sannolikheten för skott. Slutsatsen blev att det är väsentligt att tillämpa en hierarkisk modellstruktur. Av vald modell så dras slutsatsen att det som påverkar sannolikheten att komma till skott allra mest, med avseende de variabler som undersökts, är antalet händelser som sker i hörnsekvensen. Den hörnsekvenstyp som påverkas mest av antalet händelser är den korta varianten. Det diskuteras om hur det kan vara problematiskt att undersöka den linjära påverkan på log-oddset. Detta eftersom påverkan på sannolikheten för skott inte är densamma för en ökning mellan en och två händelser som mellan tre och fyra händelser. Det är även näst intill omöjligt att komma till skott på första händelsen i hörnsekvensen då händelsen utgörs av att hörnan slås. / Sports analysis is defined by Alamar (2013) as the management of structured historical data, the application of analytical models that utilize that data, and the use of information systems to inform decision makers and enable them to help their organization in gaining a competitive advantage on the field of play. This study focuses on sports analysis, more specifically corner sequences in Swedish elite football. A corner sequence is defined as the sequence of events that occur after the ball have been put into play from the corners start position up until that one of the following conditions are met: 8 events occur given a short corner is played 6 events occur given a long corner is played 15 seconds passes The defending team overtake the ball Some team performs a foul The attacking team performs a shot The data set used comes from Wyscout and is provided by Football Analytics Sweden AB. The data consist of games from the top Swedish football leagues for men: Allsvenskan and Superettan, and consists of games played in the seasons 2017, 2018 and 2019. In the data, information about every event that occur during the game is provided, where all events are classified to provide information about what happens at the specific event. The information about each event and its coordinates is then used produce variables to describe what occurs during a corner sequence. The purpose is to identify corner sequences with similar characteristics and group them together. Then use these groups to examine the probability that a corner sequence leads to a shot, and what influences this probability. The clustering algorithm Partitioning Around Medoids (PAM) is used with Gower as the dissimilarity measure and silhouette to evaluate the clusters, then the five following clusters are identified: Corners curled away from goal from the left corner with a tendency towards the front post and relatively close to the goal line. Corners curled towards goal from the left corner with a tendency towards the front post further away from the goal line. Corner curled away from goal from the right corner with a tendency towards the goal line. Short corner variant with longer duration, more events occurring and more players involved. Corners curled towards goal from the right corner with a tendency towards the front post and further away from the goal line. Given the clustering structure it is noted that the corner sequence of type 4 has led to the greatest extent of shots with the proportion of 19 percent within the cluster. The corner sequences of type 2 and 5 have a slightly lower shot occurrence with 18 percent per corner seqence type. With these corner sequence types in focus, several Hierarchical Bayesian Logistic Regression models are fitted to analyze the probability that a corner sequence leads to a shot given the produced explanatory variables. When fitting the models it is examined if it is necessary to apply a hierarchichal strutcture to the model. The conclusion is drawn that the hierarchical model structure is crucial to the model's performance. The conclusion is drawn from the final model that the explanatory variable which explains the probability to shoot best is the number of events that occur during the corner sequence. The corner sequence type that is most influenced by the number of events that occur during the corner sequence is the short corner variant. In the study it is discussed if there is an issue to suppose that this variable has a linear effect on the log-odds, since the impact on the probability to shoot is not the same for an increase between one and two as three and four events. Furthermore it is near impossible to shoot in the first event that occurs in the corner sequence.

This masters thesis has a few objectives. One of them is to acquaint with the problems of genome engineering, especially with fragmentation of DNA, the macromolecule DNA, the methods for purification and separation of the nucleic acids, the enzymes used for modification of these acids, amplification and get to know with cluster and gradient analysis as well. The next aim is to peruse the existed application and compare it to the layout of the proposed application, that is the third aim. The last one from the objectives is the implementation and the report how was the application tested by the real data. The results will be discussed as well as the possibilities of the further extension.

No description available.

No description available.

Die kumulative Promotion befasst sich mit der Lebensform „Living apart together“ (LAT), worunter Paare verstanden werden, die nach Maßgabe ihre Selbstwahrnehmung in getrennten Haushalten leben. Die Arbeit besteht aus insgesamt vier Beiträgen, die in peer-review-Journals erschienen sind sowie einer zusammenfassenden Synopse. Ein erster Schwerpunkt des Promotionsprojektes besteht darin, die Binnendifferenzierung der partnerschaftlichen Lebensform LAT zu untersuchen. Dazu werden mit Daten des Beziehungs- und Familienpanels (Befragte zwischen 15 und 39 Jahre) verschiedene Typen von LAT-Partnerschaften mithilfe von Clusteranalysen identifiziert. Hierbei erweist sich eine heuristische Einteilung in drei Idealtypen – die LAT als Vorstufe stärker verfestigter Lebensformen, LAT als berufsbedingte Notlösung und LAT als Beziehungsideal – als weitgehend empirisch tragfähig. Gleichzeitig werden neue Typen wie z.B. die „ökonomisch deprivierte LAT“ identifiziert, bei der eine ökonomisch prekäre Lage und eine starke Betroffenheit von Arbeitslosigkeit die weitere Institutionalisierung der Partnerschaft zu hemmen scheinen. Das zweite Ziel der Arbeit besteht darin, die Entwicklung der LAT-Partnerschaften im Längsschnitt, d.h. die Übergänge in den gemeinsamen Haushalt einerseits und in eine Trennung andererseits, zu untersuchen. Hier zeigen sich zum Teil deutlich Unterschiede zwischen den zuvor identifizierten Clustern. Eine niedrige Übergangsrate in die Kohabitation sowie ein hohes Trennungsrisiko können z.B. für jugendliche LAT-Partnerschaften, aber auch für den ökonomisch deprivierten Typ, beobachtet werden. Die Kohabitationsneigung bei berufsbedingten Fernbeziehungen ist dagegen höher und das Trennungsrisiko geringer als theoretisch erwartet. Schließlich wird als dritter Schwerpunkt des Projektes der Frage nachgegangen, welche Rolle Eigenschaften der Herkunftsfamilie im Institutionalisierungsprozess spielen. Es zeigt sich, dass Jugendliche die für sie typische LAT-Partnerschaft insbesondere dann früh verlassen und einen Haushalt mit ihrem Partner gründen, wenn es sich nicht um Kernfamilien, sondern um alleinerziehende Eltern oder Stiefeltern handelt. Im Promotionsprojekt wird der Frage nachgegangen, auf welche Mechanismen diese Zusammenhänge hauptsächlich zurückführbar ist, wobei verschiedene theoretische Ansätze – ökonomische Deprivation, Transmissionseffekte, soziale Kontrolle, Stress – vergleichend gegenübergestellt werden.:1. Synopse 2. Lois, Nadia 2012: "Living apart together": Sechs Typen einer heterogenen Lebensform. In: Zeitschrift für Familienforschung 24: 247-268. 3. Lois, Daniel; Lois, Nadia 2012: "Living apart together" – eine dauerhafte Alternative? Zur Bedeutung von beruflichen Lagen und Partnerschaftsbildern für das Leben in getrennten Haushalten. In: Soziale Welt 63: 117-140. 4. Lois, Nadia 2014: Einflüsse der Herkunftsfamilie auf den frühzeitigen Auszug aus dem Elternhaus und die Kohabitation – Ein Test vermittelnder Mechanismen. In: Zeitschrift für Soziologie der Erziehung und Sozialisation 34: 71-88. 5. Arránz Becker, Oliver; Salzburger, Veronika; Lois, Nadia; Nauck, Bernhard 2013: What narrows the stepgap? Closeness between parents and adult (step)children in Germany. In: Journal of Marriage and Family 75: 1130-1148.

Indiana University-Purdue University Indianapolis (IUPUI) / The aim of this study was to use multivariate statistical techniques to: (1) determine the reproducibility of fiber evidence analyzed by MSP, (2) determine whether XRF is an appropriate technique for forensic tape analysis, and (3) determine if DART/MS is an appropriate technique for forensic tape analysis. This was achieved by employing several multivariate statistical techniques including agglomerative hierarchical clustering, principal component analysis, discriminant analysis, and analysis of variance. First, twelve dyed textile fibers were analyzed by UV-Visible MSP. This analysis included an inter-laboratory study, external validations, differing preprocessing techniques, and color coordinates. The inter-laboratory study showed no statistically significant difference between the different instruments. The external validations had overall acceptable results. Using first derivatives as a preprocessing technique and color coordinates to define color did not result in any additional information. Next, the tape backings of thirty-three brands were analyzed by XRF. After chemometric analysis it was concluded that the 3M tapes with black adhesive can be classified by brand except for Super 33+ (Cold Weather) and Super 88. The colorless adhesive tapes were separated into two large groups which were correlated with the presence of aluminosilicate filler. Overall, no additional discrimination was seen by using XRF compared to the traditional instrumentation for tape analysis previously published. Lastly, the backings of eighty-nine brands of tape were analyzed by DART/MS. The analysis of the black adhesive tapes showed that again discrimination between brands is possible except for Super 33+ and Super 88. However, now Tartan and Temflex have become indistinguishable. The colorless adhesive tapes again were more or less indistinguishable from one another with the exception of Tuff Hand Tool, Qualpack, and a roll of 3M Tartan, which were found to be unique. It cannot be determined if additional discrimination was achieved with DART/MS because the multivariate statistical techniques have not been applied to the other instrumental techniques used during tape analysis.

Indiana University-Purdue University Indianapolis (IUPUI) / As medical data continues to transition to electronic formats, opportunities arise for researchers to use this microdata to discover patterns and increase knowledge that can improve patient care. Now more than ever, it is critical to protect the identities of the patients contained in these databases. Even after removing obvious “identifier” attributes, such as social security numbers or first and last names, that clearly identify a specific person, it is possible to join “quasi-identifier” attributes from two or more publicly available databases to identify individuals. K-anonymity is an approach that has been used to ensure that no one individual can be distinguished within a group of at least k individuals. However, the majority of the proposed approaches implementing k-anonymity have focused on improving the efficiency of algorithms implementing k-anonymity; less emphasis has been put towards ensuring the “utility” of anonymized data from a researchers’ perspective. We propose a new data utility measurement, called the research value (RV), which extends existing utility measurements by employing data constraints rules that are designed to improve the effectiveness of queries against the anonymized data. To anonymize a given raw dataset, two algorithms are proposed that use predefined generalizations provided by the data content expert and their corresponding research values to assess an attribute’s data utility as it is generalizing the data to ensure k-anonymity. In addition, an automated algorithm is presented that uses clustering and the RV to anonymize the dataset. All of the proposed algorithms scale efficiently when the number of attributes in a dataset is large.