In today’s society we are surrounded by large amounts of information, quick decisions and high expectations to perform successfully in everything we do. As a statistical agency, Statistics Sweden is responsible for producing some of the information that is used for decision-making in society and is therefore under constant internal and external pressure to perform well. The responsibility to produce high-quality statistics to all customers and users is not simple. What is the quality of the statistics produced? How do we assure and control the quality of the statistics? Do we use our resources efficiently? These are important questions, which need to be addressed. One way of addressing these and other issues is to work with quality in a systematic fashion. Thus there is a need for a Quality Management System, i.e., a systematic way to handle quality issues of all kinds in all parts of the organization, and to continue the journey towards the vision of being a “world class statistical agency”. This Masters thesis is a description and discussion of the efforts to find a suitable Quality Management System. The thesis starts with a discussion about the vague quality concept, continues with a description of numerous frameworks, methods and systems related to quality management as well as their pros and cons and ends with a recommendation for Statistics Sweden. The recommendation is to use the EFQM Excellence Model as a quality framework, Six Sigma as a tool-box for improvement projects and modern internal auditing methods for evaluation and follow-up. Finally, issues related to the implementation of the system are discussed.

<p>In today’s society we are surrounded by large amounts of information, quick decisions and high expectations to perform successfully in everything we do. As a statistical agency, Statistics Sweden is responsible for producing some of the information that is used for decision-making in society and is therefore under constant internal and external pressure to perform well. The responsibility to produce high-quality statistics to all customers and users is not simple. What is the quality of the statistics produced? How do we assure and control the quality of the statistics? Do we use our resources efficiently? These are important questions, which need to be addressed.</p><p>One way of addressing these and other issues is to work with quality in a systematic fashion. Thus there is a need for a Quality Management System, i.e., a systematic way to handle quality issues of all kinds in all parts of the organization, and to continue the journey towards the vision of being a “world class statistical agency”.</p><p>This Masters thesis is a description and discussion of the efforts to find a suitable Quality Management System. The thesis starts with a discussion about the vague quality concept, continues with a description of numerous frameworks, methods and systems related to quality management as well as their pros and cons and ends with a recommendation for Statistics Sweden. The recommendation is to use the EFQM Excellence Model as a quality framework, Six Sigma as a tool-box for improvement projects and modern internal auditing methods for evaluation and follow-up. Finally, issues related to the implementation of the system are discussed.</p>

I det här masterarbetet har jag musik för tre olika musikaliska sammanhang; elektronisk pop i samarbete med kurskollega, kortfilmsmusik i samarbete med filmregissör och musik till statistik i samarbete med SCB. De tre delprojekten rör sig inom olika musikaliska uttryck och har olika instrumentering. En frågeställning som genomsyrar tankegångarna i textens reflektioner är vilka likheter i fråga om uttryck och processer de tre olika verken har även om de har olika estetiska ramar och förutsättningar. Arbetet har genomförts i syfte att undersöka hur de olika inriktningarna påverkar och influerar mig som musikskapare. Går det att se hur flödet av influenser ser ut? Vidare tar jag upp olika föreställningar om att som enskild musikskapare arbeta inom olika inriktningar och medieuttryck. Metoden har utgjorts av kompositionsarbete, improvisation, musikproduktion och studioinspelningar. Resultatet av arbetet är tre olika musikverk varav två; kortfilmsmusik och musik till statistik, är en sorts beställningsverk med målet att spelas i den offentliga kulturen. / In this master dissertation, I have created music for three different musical contexts; electronic pop in collaboration with a program colleague from KMH, short film music in collaboration with film director and finally music to statistics in collaboration with Statistics Sweden (SCB. The three sub-projects move within different musical expressions and have different instrumentation. An issue that is central in the reflections of the text is what similarities in terms of expression and processes the three different works have, although they have different aesthetic frameworks. The work aims to find out how the different directions affect and influence me as a music creator. Can you see what the flow of influences looks like? Furthermore, I discuss different conceptions of working as an individual music creator in different directions and media expressions. The method has consisted of composition work, improvisation, music production and studio recordings. The result of the work is three different pieces of music, two of which; short-film music and music for statistics, is a kind of commissioned work with the goal of being played in the public culture.

The objective of this thesis is to investigate the possibility of using machine learn- ing at Statistics Sweden within the Foreign Trade in Services (FTS) statistic, to predict the likelihood of a unit to conduct foreign trade in services. The FTS survey is a sample survey, for which there is no natural frame to sample from. Therefore, prior to sampling a frame is manually constructed each year, starting with a register of all Swedish companies and agencies and in a rule- based manner narrowing it down to contain only what is classified as units likely to trade in services during the year to come. An automatic procedure that would enable reliable predictions is requested. To this end, three different machine learning methods have been analyzed, two rule- based methods (random forest and extreme gradient boosting) and one distance- based method (k nearest neighbors). The models arising from these methods are trained and tested on historically sampled units, for which it is known whether they did trade or not. The results indicate that the two rule-based methods perform well in classifying likely traders. The random forest model is better at finding traders, while the extreme gradient boosting model is better at finding non-traders. The results also indicate interesting patterns when studying different metrics for the models. The results also indicate that when training the rule-based models, the year in which the training data was sampled needs to be taken into account. This entails that cross-validation with random folds should not be used, but rather grouped cross-validation based on year. By including a feature that mirror the state of the economy, the model can adapt its rules to this, meaning that the rules learned on training data can be extended to years beyond training data. Based on the observed results, the final recommendation is to further develop and investigate the performance of the random forest model.