Stein's method is a powerful technique that can be used to obtain bounds for approximation errors in a weak convergence setting. The method has been used to obtain approximation results for a number of distributions, such as the normal, Poisson and Gamma distributions. A major strength of the method is that it is often relatively straightforward to apply it to problems involving dependent random variables. In this thesis, we consider the adaptation of Stein's method to the class of Variance-Gamma distributions. We obtain a Stein equation for the Variance-Gamma distributions. Uniform bounds for the solution of the Symmetric Variance-Gamma Stein equation and its first four derivatives are given in terms of the supremum norms of derivatives of the test function. New formulas and inequalities for modified Bessel functions are obtained, which allow us to obtain these bounds. We then use local approach couplings to obtain bounds on the error in approximating two asymptotically Variance-Gamma distributed statistics by their limiting distribution. In both cases, we obtain a convergence rate of order n^-1 for suitably smooth test functions. The product of two normal random variables has a Variance-Gamma distribution and this leads us to consider the development of Stein's method to the product of r independent mean-zero normal random variables. An elegant Stein equation is obtained, which motivates a generalisation of the zero bias transformation. This new transformation has a number of interesting properties, which we exploit to prove some limit theorems for statistics that are asymptotically distributed as the product of two central normal distributions. The Variance-Gamma and Product Normal distributions arise as functions of the multivariate normal distribution. We end this thesis by demonstrating how the multivariate normal Stein equation can be used to prove limit theorems for statistics that are asymptotically distributed as a function of the multivariate normal distribution. We establish some sufficient conditions for convergence rates to be of order n^-1 for smooth test functions, and thus faster than the O(n^-1/2) rate that would arise from the Berry-Esseen Theorem. We apply the multivariate normal Stein equation approach to prove Variance-Gamma and Product Normal limit theorems, and we also consider an application to Friedman's X² statistic.

This current study addresses potential impediments to attachment that may differentiate between incarcerated juveniles who have committed sexual crimes and incarcerated juveniles who have not committed such offenses. The exploration of such potential barriers to attachment has been organized around Bronfenbrenner's Ecological Model. Subjects were 2948 incarcerated male adolescents adjudicated to the Virginia Department of Juvenile Justice and were divided based solely upon adjudicating offense: Juvenile non-violent, non-sexual offenders (JNVNSO, n=1149), Juvenile violent, non-sexual offenders (JVNSO, n=1433) and Juvenile sexual offenders (JSO, n=366). Results indicated that JSOs differed from JNVNSOs and JVNSOs in their histories of sexual abuse as well as placements in foster care. Attachment impairment and the number of risk factors present were also found to be significantly related. Future research directions and potential policy repercussions are also addressed.

Platsannonser är medveten kommunikation från arbetsgivare med avsikten att framförallt locka nya medarbetare, men fungerar även som marknadsföring av den rekryterande organisationen. I platsannonser är efterfrågan på personliga egenskaper hos den som ska anställas ett vanligt förekommande fenomen. Dessa egenskaper är av intresse i denna studie. ”Personliga egenskaper i platsannonser” är en kvantitativ innehållsanalysstudie av platsannonser från media och kommunikationsområdet där förekomsten av efterfrågade personliga egenskaper är vad som analyseras. Studien syftar till att belysa hur platsannonser kan ses som strategiska i sin kommunikation i form av att efterfråga personliga egenskaper. Hur denna efterfråga ser ut och om det finns någon skillnad mellan storstad och landsbygd i Sverige kommer utredas. För att ta reda på vilka personliga egenskaper som efterfrågas oftast utformades ett kodningsschema och 654 platsannonser samlades in från Arbetsförmedlingen jobbförmedlartjänst Platsbanken. De 20 mest efterfrågade egenskaperna från dessa genomgick sedan ett chi2-test för att se om det fanns någon statistisk signifikant skillnad mellan kommuner i storstad och landsbygd eftersom storlek på kommun kan ha betydelse för hur arbetet utförs. Teorierna som studien baseras på är strategisk kommunikation och femfaktorteorin eller big five teorin som den heter som den heter på engelska. Dessa teorier visar på varför en platsannons kan ses som strategisk kommunikation och hur personlighet utgör en aspekt i rekryteringsprocessen. Femfaktorteorin bygger på att människans personlighet bygger på de fem faktorerna. Dessa faktorer består av adjektiv som beskriver personlighet, såsom t.ex. ansvarsfull, nyfiken och noggrann Alla människor har mer eller mindre av dessa faktorer. Studiens resultat visar att vissa personliga egenskaper efterfrågas mer frekvent än andra och från ett femfaktorteoretiskt perspektiv är vissa av dess faktorer mer förekommande än andra. Vilket i sin tur pekar på att platsannonser är strategisk kommunikation. Det fanns skillnader mellan storstad och landsbygd i vilka egenskaper som efterfrågas men likheterna var fler än olikheterna. Den vanligaste personliga egenskapen som efterfrågades i det insamlade materialet var strukturerad. Denna egenskap tillhör personlighetsfaktorn samvetsgrannhet. Att inneha mer av faktorn samvetsgrannhet är en bra indikation på framtida yrkesmässiga lycka, nöjdhet och prestation.

The main research objective of this thesis is to determine if, or how, the verbal communication of a youth football coach differs from the coaches at the club level of the regional competitions and republic competitions. The theoretical section of the thesis deals with coaching and communication. The first chapter describes the work of a football coach as an educator. The second chapter, focused on communication, further describes concepts such as verbal, pedagogical and effective communication. The empirical section presents the results of a quantitative survey. It has been shown that coaches' communica- tion with players varies significantly, for example, in the content of commands, remorse, vulgarism, or in frequency of shouting towards a player after his mistake. The frequency of occurrence of the above-mentioned communication features is higher for coaches at regional club level.

Med en explorativ ansats syftar denna studie till att kartlägga samband och jämföra grupperingar bland klienter som fått genomföra en ASI-intervju hos Kriminalvården. Studien bygger på 2 317 intervjuer med fokus på missbruksproblematik, gjorda under perioden 2008-2010. Vi illustrerar grafiskt intressanta samband och tittar även på hur klienternas hjälpbehov varierar inom olika livsområden. Ytterligare en aspekt av studien är att utvärdera datamaterialets kvalitet samt att studera eventuellt bortfall för de olika variablerna. Materialet är av relativt god kvalitet och förekomst av respondenters vägran är inte tillräckligt omfattande för att påverka analyser och slutsatser. De felaktigheter som kunde konstateras härhör nästan uteslutande från tillvägagångssätt vid inmatning av data i systemet. Eftersom flertalet av variablerna är kvalitativa visade sig lämpliga analysmetoder vara chi-två-test, korrespondensanalys samt associationsanalys. Den stora majoriteten av våra resultat bekräftar den bild flera tidigare studier ger av gruppen missbrukare och få oväntade samband kunde konstateras. En majoritet av klienterna är män (85%) och den genomsnittliga åldern är 34 år. Intervjuarskattningar av klienternas hjälpbehov visar att narkotika och kriminalitet utgör de absolut största problemen. Dominerande drog är amfetamin (18%), följd av alkohol (15%) och cannabis (10%). Vi kan konstatera att ett flertal signifikanta skillnader föreligger mellan män och kvinnor samt mellan olika åldersgrupper. Unga klienter uppvisar överlag större problem med kriminalitet, narkotika, arbete/försörjning och psykisk hälsa. Med ökande ålder följer istället svårigheter med fysisk hälsa och alkohol. Män uppger vanligen att de har större problem med kriminalitet och narkotika, medan kvinnor ofta redovisar ett större hjälpbehov inom områdena familj/umgänge, fysisk hälsa och psykisk hälsa. / With an explorative approach, this master thesis attempts to map associations and compare groupings of clients who have undergone an ASI Interview by The Swedish Prison and Probation Service, Kriminalvården. The study is based on 2 317 interviews carried out during the period 2008-2010, focusing on abuse of alcohol and narcotic substances. We will graphically illustrate interesting associations and study how the need for help varies throughout different areas of the respondents’ lives. Another aspect of this study is to evaluate the quality of the data material and to investigate non-response in the different variables. The quality of the material is fairly high and non-response is not extensive enough to affect analysis and inference. The errors found are almost exclusively due to how data is fed into the system. Since the majority of variables are qualitative, appropriate methods of analysis proved to be chi-square tests, correspondence analysis and association analysis. Most results confirm the existing image of an abusive personality, which has been presented in several studies before this one. A majority of the clients are men (85%) and the average age is 34 years. Interviewer estimates of the respondents’ need for help show that narcotic substances and criminal behavior are by far the greatest problem areas. The predominant drug is amphetamine (18%), followed by alcohol (15%) and cannabis (10%). We have established several significant differences between men and women, and between clients of different age groups. Young respondents show greater problems concerning criminal behavior, narcotic substances, work/providing and psychological health. With increased age we see enhanced difficulties regarding physical health and alcohol. Men more often exhibit problems concerning criminal behavior and narcotic substances, while women need increased help in areas of family/social life, physical health and psychological health.

The research is dealing with the problem of choosing right statistical criteria in order to analyze the concrete educological research information and to interpret the received results. The practical value of the research’s results could be described as: 1- the exact evaluation of the real situation, by researching the effectiveness of use of educational computer means at the lessons of mathematics, with the help of hypothesises; 2- the statistical criteria received at the time of the research can be used at the concrete future researches.

Made available in DSpace on 2014-12-17T14:06:51Z (GMT). No. of bitstreams: 1 AntonioEFF.pdf: 720128 bytes, checksum: febce55b295b3d3be15f1c24bf498884 (MD5) Previous issue date: 2009-02-26 / Na unfolding method of linear intercept distributions and secction ?rea distribution was implemented for structures with spherical grains. Although the unfolding routine depends on the grain shape, structures with spheroidal grains can also be treated by this routine. Grains of non-spheroidal shape can be treated only as approximation. A software was developed with two parts. The first part calculates the probability matrix. The second part uses this matrix and minimizes the chi-square. The results are presented with any number of size classes as required. The probability matrix was determined by means of the linear intercept and section area distributions created by computer simulation. Using curve fittings the probability matrix for spheres of any sizes could be determined. Two kinds of tests were carried out to prove the efficiency of the Technique. The theoretical tests represent ideal cases. The software was able to exactly find the proposed grain size distribution. In the second test, a structure was simulated in computer and images of its slices were used to produce the corresponding linear intercept the section area distributions. These distributions were then unfolded. This test simulates better reality. The results show deviations from the real size distribution. This deviations are caused by statistic fluctuation. The unfolding of the linear intercept distribution works perfectly, but the unfolding of section area distribution does not work due to a failure in the chi-square minimization. The minimization method uses a matrix inversion routine. The matrix generated by this procedure cannot be inverted. Other minimization method must be used / A rotina de desdobramento (unfolding) de distribui??es de intercepto linear e de ?rea de se??o foi implementada para gr?os de formato esf?rico. Apesar da rotina de unfolding ser fortemente dependente do formato do gr?o, estruturas que possuam gr?os com formatos esferoidais podem ser tratadas com esta rotina. Mesmo estruturas com gr?os de formatos n?o esferoidais podem ser tratadas por aproxima??o. Um programa com duas partes foi desenvolvido. Primeira parte determina a tabela de probabilidades. A segunda utiliza esta tabela e aplica o m?todo de minimiza??o do chi-quadrado. Os resultados s?o dados em qualquer n?mero de classes de tamanho de gr?o requerido pelo usu?rio. A tabela de probabilidade foi determinada a partir de distribui??es de intercepto linear e de ?rea de se??o geradas por simula??o computacional. Por meio de ajustes de curvas de distribui??o, tabelas de probabilidades para esferas de qualquer tamanho podem ser determinadas. Dois tipos de testes foram executados para verificar a efici?ncia do m?todo. Os testes te?ricos representam situa??es ideais. O programa conseguiu reproduzir com exatid?o as distribui??es de tamanho de gr?o sugeridas. Os testes simulados consistem em simular em computador distribui??es de tamanho de gr?os e executar todo o procedimento metalogr?fico usual. Este tipo de teste ? mais pr?ximo da situa??o real. Os resultados deste tipo de teste mostram que a rotina de medi??o estereol?gica introduz desvios estat?sticos, afastando o resultado encontrado do valor real. Contudo, a rotina de unfolding funciona perfeitamente para a distribui??o de intercepto linear. No caso de unfolding de ?rea de se??o, a minimiza??o do chi-quadrado pelo m?todo matricial gera matrizes n?o invers?veis e n?o pode ser aplicada. Outro m?todo de minimiza??o deve ser buscado

The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 has lead to reduced welfare assistance to the needy. This thesis analyzes how families have become increasingly homeless and hungry during the welfare reform years.

he point of this diploma thesis is to analyze current marketing strategies of Strasil Co. and create marketing quantitative research. Based on the obtained data, author is trying to develop the concept of solutions, which will increase the volume of new customers.

This master's thesis presents methods for mining important pieces of information from text. It analyses the problem of terms extraction from large document collection and describes the implementation using C# language and Microsoft SQL Server. The system uses stemming and a number of statistical methods for term extraction. This project also compares used methods and suggests the process of the dictionary derivation.