På sistone har allt mer kritik riktats mot forskning inom klassificering. Trots att forskningen har resulterat i en uppsjö av klassificeringsmetoder finns det de som menar att den har varit ett misslyckande och pekar på det faktum att ingen klassificeringsmetod anses vara systematiskt bättre än den andra eller ens rena gissningar. Detta leder till att valet av klassificeringsmetod i många fall vilar på individuella preferenser snarare än på vetenskaplig grund. Enligt litteraturen bottnar detta faktum i ett underliggande samband mellan det optimala valet av klassificeringsmetod och egenskaperna som karaktäriserar datamaterialet.Uppsatsen tar avstamp från denna problematik och syftar till att undersöka kopplingen mellan det optimala valet av klassificeringsmetod och datamaterialets egenskaper. Denna intention uppnår vi genom att tillämpa logistisk regression, elastic net och boosting tillämpat på klassificeringsträd på sex verkliga datamaterial med varierande statistiska egenskaper.Resultatet visar att den relativa klassificeringsförmågan varierar med datamaterialen.

Today wireless transmission of data is becoming more and more popular and the need for faster transmission rates is increasing. Since the bandwidth is limited it is important to try to use it to the fullest. CDMA is a technology that allows multiple accesses on the same frequency and time thus making it very bandwidth efficient. The CDMA technology was first introduced in the second generation?s cellular systems but has since then been improved and is reused in today?s 3G systems as Wideband CDMA.

The Swedish Transport Administration has, and manages, a database containing information of the status of road condition on all paved and governmental operated Swedish roads. The purpose of the database is to support the Pavement Management System (PMS). The PMS is used to identify sections of roads where there is a need for treatment, how to allocate resources and to get a general picture of the state of the road network condition. All major treatments should be reported which has not always been done.The road condition is measured using a number of indicators on e.g. the roads unevenness.

The task given to us was to develop an industrial wheel with new thinking in terms of fresh design and construction, with the ability to brake. While the produced wheels of today only own the property of total braking, a wish was made to integrate the ability of directional locking.The time spent working with the project has been filled with multiple sketches of ideas and function analysis.Choices of material and ways of construction have been verified by calculations, tests and prototype building. The goal of this thesis has been to develop a unique caster with well thought-out design that would be adaptable to different areas of applications.The result was a hubless wheel with the ability of directional locking before total braking.A prototype of the wheel has been made to show the design and functions..

This essay covers computer gaming as seen from a perspective of gender theory. The main purpose of the essay is to examine women?s relationship to computer games, and more specifically women?s attitudes to the games they play. The theoretical basis for this essay is sprung from a hermeneutic perspective. The empirical basis for analysis consists of two in depth interviews and game analysis of the two computer games Counter Strike and Final Fantasy XI.

In this essay we have approached the problem with retailer brands that copy marketing leaders to get consumers attention in the storeshelf. We have sought to explore how much a designed package would change the consumers opinion of a brand and if this effect is different depending on if the brand is a retailer brand or a market leader. Also we sought to find if this effect would be greater with a salient color on the package. We made a quantitative study and with regression analysis we found that a designed package did have effect on the consumers opinion of the brand, and that the design with the salient color made the biggest effect..

The aim of this study was to evaluate different types of volume and taper equations that can be used to predict single-tree stem volume and stem diameter at any given height along the tree stem for plantation grown Sitka spruce (Picea sitchensis (Bong.) Carr.), Norway spruce (Picea abies (L.) Karst) and White spruce (Picea glauca (Mounch) Voss) in Iceland. A number of published tree volume equations were tested and modified to predict the total stem volumes over bark but three logarithmic equations were taken for more in-depth analysis. Three taper equations were tested. Two variable-exponent equations (Kozak 1997, Kozak 2004) and one exponential equation described by Biging (1984). Data from a total of 617 sample trees were used in this study, collected from stands in various parts of the country and present different types of stands growing in different soil types and cover most of the site conditions suitable for forestry in Iceland.

AimThe purpose of this study is to analyze factors that affect the price on tenant-owner apartments in the central parts of Uppsala. Special attention is put on analyzing how the monthly fee and the location affect the price. The hypotheses are that the monthly fee and the distance to the central part of the city have a negative effect on the price. A number of additional price affecting factors was taken in consideration during this study.MethodThis study is mainly based on data supplied to us by Mäklarstatistik. The supplied data consists of information about tenant-owner apartment sales in Uppsala during a year under 2008 and 2009.

The wide distribution range of our boreal pine (Pinus sylvestris L.) provides a great variation in growing conditions and climate/weather factors that influence the wood development of the trees. A great number of extensive studies have been conducted over the years in order to clarify growth trends of this and other species over more or less wide-ranging areas. In this thesis effects of weather factors were instead compared between different forest stands from the about same geographical location. The effects of precipitation, temperature and frost on the year ring characteristics of pine were studied in the research forest of Svartberget Vindeln, in northern Sweden. The stands differed in site index and silvicultural treatment; fertile ? high stem density, fertile ? low stem density, poor ? high stem density and poor ? low stem density.

År 2007 är det högkonjunktur och arbetslösheten sjunker i Sverige. Fler väljer att läsa vidare på högskola och universitet och färre människor föds. En attraktiv arbetsplats vinner konkurrensen om den kompetens som finns. Arbetslivsinstitutet startade därför, tillsammans med Högskolan Dalarna, ett gemensamt forskningsprojekt 2001 som avslutades 2003. Projektet resulterade bla.

In a time when economic crises relieving one another, it is especially interesting to study how companies choose to finance their operations. Since there is a lack of studies that include banks, and their capital structures, with the rest of the finance market, it is uncertain whether the theories of capital structure are applicable. Therefore, this essay aims to study whether there is any correlation between the two variables debt-equity ratio and return on equity (ROE) for Swedish corporations on the finance market.In accordance with the theorems that the professors Modigliani and Miller filed, in 1958 and 1961, which describes a clear link between debts, increase corporate value through the tax shield and indirectly increased yields. Although these theorems are the foundation of the modern research they focus on perfect markets. Therefore, this paper has also chosen to place great emphasis on theories such as Trade-off and Pecking-order.

Practitioners of financial statement analysis are dependent on key ratios in a multiple number of situations, one in particular being equity valuation. A lack of historical benchmarks, and knowledge regarding the time series behaviour of such ratios, has been identified regarding Swedish companies. Therefore, this study sets out to provide such data and knowledge. Companies listed on the Stockholm Stock Exchange are examined over the period 1979-2009 where the sample is divided into nine industry groups. Historical values for eight common key ratios are documented with regards to median, mean, standard deviation and percentiles.

In this study, the effects of using fine milled cassava stems as an additive in biofuel pellet production was compared to the effects of refined starch addition. The bulk biomass fuel raw material, to which the additive was added, was a blend of spruce and pine sawdust. An experimental design in the factors cassava/starch content, moisture content and material temperature was used. Measured responses were pellet bulk density, pellet durability, amount of fines, pelletizer motor current, pellet temperature, die temperature and CV for pelletizer motor current (a measure of process stability). Each response was modeled by multiple linear regression (MLR).

Abstract Titel: Det svenska konsumtionsbeteendet ? En ekonometrisk analys av den permanenta inkomsthypotesenTopic/course: NEKK01, Bachelor thesisWriters: Ida Hedlund, Anna Persdotter Supervisor: Fredrik NG Andersson Keywords: Permanent income hypothesis, Milton Friedman, Consumption smoothing, Sweden Purpose: The purpose of this thesis is to examine whether the permanent income hypothesis can be applied on the consumption pattern of Sweden during the time period 1950-2006. Method: To do the observation an econometric regression analysis has been made. The variables tested are consumption, real GDP, stock- and real estate market prices. In the model three different time periods have been investigated to facilitate the origin of different patterns.

Betting under ordered forms has been around for a long time, but the recent increase in Internet betting and the large sums of money that are now involved makes it even more important for betting companies to have correct odds. The purpose of the essay is to calculate probabilities for outcomes of football games using a statistical model and to see if you can find better odds than a betting company.The data contains the 380 games from the 2004/2005 season and the variables form, head-to-heads, league position, points, home/away, average attendance, promoted team, distance and final league position from previous season. After performing an ordered probit regression we only find the variable ?form of the away team? to be significant at the 5 % level. We suspect the presence of multicollinearity and perform a VIF-test which confirms this. To fix this problem we perform a second ordered probit regression where a number of variables are combined to index variables. In the second regression we once again find only one significant variable.