Research question: This thesis analyzes the relationship between Altman’s Z’-and Z’’ score model in order to investigate the suitability of the models on the swedish small and medium companies during 2015-2019. Furthermore, analysis of previous research key figures has been examined in more detail to identify which of the individual key figures can be categorized as an early warning signal. Purpose: The purpose with this study is to explore which and when early warning signals can be read in annual reports between inactive and active Swedish tourist companies and to investigate the relationship between the accuracy of bankruptcy prediction models between active and bankrupt companies. Method: The study uses a quantitative method with a deductive approach, z-test and onesided analysis of variance ANOVA to analyze the accuracy of bankruptcy prediction models and identify how the key figures differ between active and inactive companies. Conclusion: The study shows that Altman's Z 'and Z' scores predict bankruptcies better than specify continued operations, are best suited for active companies and have the highest accuracy one year in advance than a longer period. On the other hand, 7 out of 15 key figures examined have identified significant average value differences between active and bankrupt companies, where some had a higher value for bankrupt companies and others had lower ones. / Problemställning: I denna studien har relationssamband mellan Altmans Z’- och Z’’- scoremodell analyserats för att undersöka hur modellerna lämpar sig på den svenska små- och medelstora konkur-respektive aktiva turistföretag mellan 2015-2019. Ytterligare har analys av tidigare forsknings nyckeltal undersökts närmare för att identifiera vilka av de enskilda nyckeltalen kan kategoriseras som en tidig varningssignal. Syfte: Studiens avsikt är att utforska vilka och när tidiga varningssignaler kan utläsas i årsredovisningar mellan inaktiva och aktiva svenska turistföretag samt undersöka relationssambandet gällande konkursprediktionsmodellers träffsäkerhet mellan aktiva- och konkursföretag. Metod: I studien används en kvantitativ metod med en deduktiv ansats, z-test och ensidig variansanalys ANOVA för att analysera konkursprediktionsmodellers träffsäkerhet samt identifiera hur nyckeltalen skiljer sig mellan aktiva och konkursföretag. Slutsats: Studien visar att Altmans Z”och Z’-score förutser konkurser bättre än preciserar fortsatt verksamhet, lämpar sig bäst på aktiva företag samt har den högsta träffsäkerhet ett år i förväg än längre period. Däremot har 7 av 15 undersökta nyckeltal identifierat signifikanta medelvärdesskillnader mellan aktiva och konkursföretag där några hade ett högre värde gällande konkursföretag och andra hade lägre.

Published Article / A study was conducted to determine the differences between four breeds in respect of coat score, hide-thickness, tick burden and body condition score. The study was comprised of 40 heifers - 10 of each breed, namely Afrikaner, Braford, Charolais and Drakensberger. A subjective system of coat scoring, ranging from extremely short to very woolly, was used. Body condition score was measured subjectively, with 1 being emaciated and 9 being obese. Hide-thickness (in mm) and tick count were also determined. Between August 2007 and early March 2008, measurements were carried out on the same 10 animals of each breed, with highly significant differences in body condition score, hide-thickness and tick count being observed between the breeds in all instances. Coat scores differed significantly between breeds in the earlier and latter stages of the study becoming less significant midway through. A significant difference in body condition score within breeds was also found, while hide-thickness did not differ significantly within breeds.

Songs of Dusk and Renewal sets two poems from Alfred Edward Housman’s A Shropshire Lad (1896) for baritone voice and orchestra, with the intent to demonstrate the timeless significance of the text. Housman’s collection, largely explored by English composers in the first half of the twentieth century, remains a valid source for composers, yet the writings are seldom observed in recent music. In the present work, poems IV (“Reveille”) and XLVI (“Bring, in this timeless grave to throw”) are reconciled musically through deep constructs and surface attraction, while being informed by critical and historical perspectives. / text

A satisfação dos clientes é fundamental para a manutenção do relacionamento com a empresa. Quando eles precisam resolver algum problema, a empresa necessita proporcionar bom atendimento e ter capacidade de resolutividade. No entanto, o atendimento massificado, muitas vezes, impossibilita soluções sensíveis às necessidades dos clientes. A metodologia estatística pode ajudar a empresa na priorização de clientes com perfil a reclamar em um órgão de defesa ao consumidor (ODC), evitando assim uma situação de atrito. Neste projeto, foi realizada a modelagem do comportamento do cliente com relação à propensão ao atrito. Foram testadas as técnicas de Regressão Logística, Random Forest e Algoritmos Genéticos. Os resultados mostraram que os Algoritmos Genéticos são uma boa opção para tornar o modelo mais simples (parcimonioso), sem perda de performance, e que o Random Forest possibilitou ganho de performance, porém torna o modelo mais complexo, tanto do ponto de vista computacional quanto prático no que tange à implantação em sistemas de produção da empresa. / Customer satisfaction is key to maintaining the relationship with the company. When these need to solve some problem, the company needs to provide good service and have resolving capacity. However, the mass service often makes it impossible. The statistical methodology can help the company in the prioritization of clients with profile to complain in ODC, thus avoiding a situation of attrition. In this project was carried out the modeling of the behavior of the client in relation to the propensity to attrition. Logistic Regression, Random Forest and Genetic Algorithms were tested. The results showed that the Genetic Algorithms are a good option to make the model simpler (parsimonious) without loss of performance and that Random Forest allowed performance gain, but it makes the model more complex, both from the point of view computational and practical in relation to the implantation in production systems of the company.

Les travaux entrepris dans cette thèse s'inscrivent dans le cadre du projet BodyScoring. Ce dernier propose une solution innovante basée sur l'utilisation d'une technologie embarquée (BodyTrack) et des applications web (BodyLink) pour évaluer les habiletés motrices et pour développer la motivation pour l’accomplissement d’une pratique physique régulière. Dans le cadre de ce projet, notre apport a consisté à évaluer et à noter la qualité de la marche des personnes âgées par le biais de capteurs inertiels qui incluent accéléromètre, gyromètre et magnétomètre. Notre apport original a consisté à caractériser le pattern de marche au travers de différentes configurations de capteurs placés sur le corps et de proposer un score global validé et facilement interprétable. Le score permettra de se positionner par rapport à une population de référence jeune et asymptomatique et in fine autoévaluer l’évolution de sa marche. Afin d’atteindre cet objectif plusieurs étapes sont nécessaires. Ainsi, le premier chapitre de ce mémoire décrit en se référant à la littérature les paramètres déterminants de la marche, les facteurs pouvant les influencer et les moyens utilisés pour les quantifier. Le second chapitre porte principalement sur la définition de la meilleure configuration de capteurs pour la détection des événements clés de la marche qui sont la survenue du contact initial et final et la quantification des paramètres temporels. Il en ressort que le gyromètre fixé au bord distal du tibia est la configuration la plus précise aussi bien pour la détection des événements de la marche que pour la quantification des paramètres temporels chez des sujets sains. Le troisième chapitre expose un nouveau protocole expérimental afin de définir les paramètres pertinents pour caractériser la marche et définir l'incidence de la pratique de la marche nordique sur les paramètres biomécaniques. Autrement dit, définir les paramètres biomécaniques qui rendent compte de l'altération du pattern de marche au cours de la sénescence ou encore apprécier l'effet d'une activité physique régulière. Cette étude a révélé 72 paramètres au pouvoir discriminant et rejoint les études qui rapportent un effet bénéfique de lamarche nordique. Pour finir, le quatrième chapitre décrit l'élaboration de nouveaux scores d'évaluation de la marche basé sur les paramètres mis en évidence au chapitre 3 complémentés par ceux qualifiant la symétrie des membres inférieurs et supérieurs. Ces derniers décrivent la qualité de la marche dans son ensemble (score global) et la qualité de chaque aspect (score partiel). Quantifiés pour trois groupes de sujets (âgés sédentaire, âgés sportif et jeune) ces scores ont permis de mettre en évidence l'altération du pattern de marche au cours de la sénescence et l'effet de la pratique d'une activité physique sur les paramètres associés à la marche. / The purpose of this thesis is to asses and scores the gait quality of elderly persons through inertial sensors. The originality of this contribution is to characterize the pattern of walking through different sensor configurations and propose an overall score, valid and easily interpretable. This latter, allows subjects to self-assess to position themselves compared to a young and asymptomatic reference population and ultimately track their evolution.The first chapter, following a review of the literature, identifies the determinant gait parameters, its influent factors and the means used to quantify them.The second chapter, focuses on the definition of the best configuration of sensors to detect gait events and quantify temporal parameters.The third chapter, lists the biomechanical parameters that reflect the changing pattern of walking during senescence or consecutive to a regular physical activity.In the fourth chapter, a new method to compute the score based on the parameters identified in Chapter 3 was developed.

<p><strong>Datum: </strong>2009-06-02</p><p><strong>Nivå: </strong>Magisteruppsats i ekonomistyrning, 15 hp</p><p><strong>Författare</strong>: Charlotta Lind och Martin Sloberg</p><p><strong>Titel: </strong>Ditt företag kan inte förutse konkurs -kan Z-score?</p><p><strong>Handledare: </strong>Esbjörn Segelod</p><p><strong>Problem: </strong>Våra forskningsfrågor är:</p><ul><li>Går det att förutse konkurs tre, fyra respektive fem år innan</li></ul><p>          konkursbeslutet?</p><ul><li>Vilka av den senaste Z-scoremodellens fyra nyckeltal är viktigast vid</li></ul><p>           prognostisering av konkurs?</p><p><strong>Syfte: </strong>Att testa i vad mån Z-scoremodellen kan användas för att förutse konkurser</p><p>bland icke börsnoterade, icke tillverkande företag tre, fyra respektive fem år</p><p>innan konkurs; samt att undersöka vilka av denna modells inbördes nyckeltal</p><p>som är viktigast vid predicering av konkurser.</p><p><strong>Metod: </strong>Vi har genom kvantitativ metod analyserat årsredovisningar från 51 företag</p><p>som gått i konkurs 2008, dessa utgjorde vår huvudundersökningsgrupp och 29</p><p>slumpmässigt utvalda företag, vilka utgjorde vår kontrollgrupp. Analysen</p><p>skedde genom användandet av Altmans vidareutvecklade modell för att</p><p>förutspå konkurser från år 1995. Totalt analyserades på detta sätt 240</p><p>årsredovisningar.</p><p><strong>Slutsats: </strong>Modellens träffsäkerhet för de undersökta konkurs företagen var</p><p>2003 45,09 %</p><p>2004 47,05 %</p><p>2005 54,90 %</p><p>Vid hypotesprövning kunde vi endast för år 2005 påvisa samband för</p><p>företagsklassificeringsfrekvenser mellan konkursföretagen och</p><p>kontrollgruppen, detta gör att modellens prognostisering bör anses alltför</p><p>osäker tidigare än tre år innan konkurs. Mot bakgrund till de påvisade</p><p>träffsäkerheterna för åren och hypotesprövningarna anser vi att modellen</p><p>endast bör användas i kombination med andra analysformer .</p><p>Sammanfattningsvis är Z-scoremodellens prognostiseringsförmåga för svag att</p><p>självständigt förutse konkurser.</p><p><strong>Sökord: </strong>Konkurs, Z-score</p> / <p><strong>Date:</strong> 2009-06-02<strong> </strong></p><p><p><strong>Level:</strong> Master thesis in Management Accounting, 15 hp</p><p><strong>Authors: </strong>Charlotta Lind and Martin Sloberg</p><p><strong>Title: </strong>Your company cannot predict bankruptcy;- can Z-score?</p><p><strong>Tutor: </strong>Esbjörn Segerlod</p><p><strong>Our problem questions:</strong></p></p><ul><li>Is it possible to predict a bankruptcy three, four or five years before</li></ul><p>           the adjudication of bankruptcy?</p><ul><li>Which one of the four keyratios in the Z-scoremodel is the most</li></ul><p>          important when predicting a bankruptcy?</p><p><p><p><strong>Purpose: </strong></p><p>To test if the Z-score model can be used to predict bankruptcy for</p></p></p><p>private own companies three, four or five years before the</p><p>adjudication. To get knowledge which one of the key ratios is most</p><p>important when predicting a bankruptcy?</p><p><p><p><strong>Method:</strong></p>Through a quantitative study of Altman's Z-score model has 51</p></p><p>bankrupt companies, 29 control companies and 240 annual reports</p><p>been analyzed.</p><p><p><p><strong>Conclusion:</strong></p>The Z-score model's accuracy for the studied bankrupt companies</p></p><p>is:</p><p>2003 45,09 %</p><p>2004 47,05 %</p><p>2005 54,90 %</p><p>Only in 2005 could a relationship between the bankrupt companies</p><p>and the control companies be established through the Z-score model</p><p>tests. This makes the model too uncertain to be used earlier than</p><p>three years before the adjudication of bankruptcy. It is therefore our</p><p>opinion that the Z-score model is too weak to be used by itself but</p><p>should rather be used as a complement with other models to predict</p><p>bankruptcies.</p><p><p><p><strong>Keywords:</strong> Bankruptcy, Z-score model</p></p></p>

Datum: 2009-06-02 Nivå: Magisteruppsats i ekonomistyrning, 15 hp Författare: Charlotta Lind och Martin Sloberg Titel: Ditt företag kan inte förutse konkurs -kan Z-score? Handledare: Esbjörn Segelod Problem: Våra forskningsfrågor är: Går det att förutse konkurs tre, fyra respektive fem år innan           konkursbeslutet? Vilka av den senaste Z-scoremodellens fyra nyckeltal är viktigast vid            prognostisering av konkurs? Syfte: Att testa i vad mån Z-scoremodellen kan användas för att förutse konkurser bland icke börsnoterade, icke tillverkande företag tre, fyra respektive fem år innan konkurs; samt att undersöka vilka av denna modells inbördes nyckeltal som är viktigast vid predicering av konkurser. Metod: Vi har genom kvantitativ metod analyserat årsredovisningar från 51 företag som gått i konkurs 2008, dessa utgjorde vår huvudundersökningsgrupp och 29 slumpmässigt utvalda företag, vilka utgjorde vår kontrollgrupp. Analysen skedde genom användandet av Altmans vidareutvecklade modell för att förutspå konkurser från år 1995. Totalt analyserades på detta sätt 240 årsredovisningar. Slutsats: Modellens träffsäkerhet för de undersökta konkurs företagen var 2003 45,09 % 2004 47,05 % 2005 54,90 % Vid hypotesprövning kunde vi endast för år 2005 påvisa samband för företagsklassificeringsfrekvenser mellan konkursföretagen och kontrollgruppen, detta gör att modellens prognostisering bör anses alltför osäker tidigare än tre år innan konkurs. Mot bakgrund till de påvisade träffsäkerheterna för åren och hypotesprövningarna anser vi att modellen endast bör användas i kombination med andra analysformer . Sammanfattningsvis är Z-scoremodellens prognostiseringsförmåga för svag att självständigt förutse konkurser. Sökord: Konkurs, Z-score / Date: 2009-06-02 Level: Master thesis in Management Accounting, 15 hp Authors: Charlotta Lind and Martin Sloberg Title: Your company cannot predict bankruptcy;- can Z-score? Tutor: Esbjörn Segerlod Our problem questions: Is it possible to predict a bankruptcy three, four or five years before            the adjudication of bankruptcy? Which one of the four keyratios in the Z-scoremodel is the most           important when predicting a bankruptcy? Purpose: To test if the Z-score model can be used to predict bankruptcy for private own companies three, four or five years before the adjudication. To get knowledge which one of the key ratios is most important when predicting a bankruptcy? Method: Through a quantitative study of Altman's Z-score model has 51 bankrupt companies, 29 control companies and 240 annual reports been analyzed. Conclusion: The Z-score model's accuracy for the studied bankrupt companies is: 2003 45,09 % 2004 47,05 % 2005 54,90 % Only in 2005 could a relationship between the bankrupt companies and the control companies be established through the Z-score model tests. This makes the model too uncertain to be used earlier than three years before the adjudication of bankruptcy. It is therefore our opinion that the Z-score model is too weak to be used by itself but should rather be used as a complement with other models to predict bankruptcies. Keywords: Bankruptcy, Z-score model

Oregon’s Kindergarten Assessment (KA) is mandatory for all incoming Oregon kindergarteners starting in the 2013-14 school year. One component of Oregon’s KA is the Child Behavioral Rating Scale (CBRS), which Oregon has adapted into the Approaches to Learning Assessment. Teachers complete the CBRS during the first four to six weeks of school. This study uses a convenience sample of 731 kindergarten students (across two years) from one district in Oregon to analyze behavioral readiness (self-regulation and social-emotional behaviors) as well as easyCBM indicators of academic readiness. The CBRS is compared with the Child Behavioral Checklist and the Ages and Stages Questionnaire: Social Emotional as criterion measures. Parent and teacher responses to the CBRS are analyzed for comparability, and a Receiver Operating Characteristic curve analysis of the data is used to determine optimal cut points (maximizing sensitivity and specificity) for predicting whether students are at risk compared to the criterion measure cut scores. Demographic variables of gender, English Language Learner status, and Socioeconomic Status, are analyzed as control variables. Pre-post behavior change on the CBRS is document over the kindergarten year, and kindergarten academic benchmark measures is used as a dependent measure. This study explores whether: (a) parent responses differ significantly from teacher responses (internal consistency), (b) a cut score on the CBRS successfully sorts students into categories of "typically developing" or "in need of further assessment," (c) teacher predictions align to the proposed CBRS cut score, (d) academic risk is correlated to the established CBRS cut score, and (e) change in behavior over the course of kindergarten is measured (pre-post) by the CBRS. Results from this research could support identification of students for interventions in both kindergarten and early childhood programs.

Equating is a statistical process that is used to adjust scores on test forms so that scores on the forms can be used interchangeably. This dissertation offered intensive investigation of beta true and observed score methods by comparing them to existing traditional and IRT equating methods under multiple designs and various conditions using real data, pseudo-test data and simulated data. Weighted and conditional bias, standard error of equating and root mean squared error were used to evaluate the accuracy of equating results obtained from the pseudo data and simulated data analyses. The single group equipercentile equating based on large sample sizes was used as the criterion equating. Overall, results showed that of the methods examined, the IRT methods performed best, followed by the chained equipercentile methods. Results from beta methods presented different trends from traditional and IRT methods for both the random group and common item nonequivalent groups designs. Beta true scores methods were less sensitive to group difference compared to traditional methods. The length of common items played an important role in the stability of results of beta true score methods.

published_or_final_version / abstract / toc / Paediatrics and Adolescent Medicine / Doctoral / Doctor of Philosophy