Considere um processo estocástico X_m em tempo discreto definido sobre o alfabeto finito A. Seja x_0^k-1 uma palavra fixa sobre A^k. No estudo das propriedades estatísticas na teoria de recorrência de Poincaré, é clássico o estudo do tempo decorrente até que a sequência fixa x_0^k-1 seja encontrada em uma realização do processo. Tipicamente, esta é uma quantidade exponencialmente grande com relação ao comprimento da palavra. Contrariamente, o primeiro tempo de retorno possível para uma sequência dada está definido como sendo o mínimo entre os tempos de entrada de todas as sequências que começam com a própria palavra e é uma quantidade tipicamente pequena, da ordem do tamanho da palavra. Neste trabalho estudamos o comportamento da probabilidade deste primeiro retorno possível de uma palavra x_0^k-1 dado que o processo começa com ela mesma. Esta quantidade mede a intensidade de que, uma vez observado um conjunto alvo, possam ser observados agrupamentos ou clusters. Provamos que, sob certas condições, a taxa de decaimento exponencial desta probabilidade converge para a entropia para quase toda a sequência quando k diverge. Apresentamos também um estimador desta probabilidade para árvores de contexto e mostramos sua consistência. / Considering a stochastic process X_m in a discrete defined time over a finite alphabet A and x_0^k-1 a fixed word over A^k. In the study of the statistical properties of the Poincaré recurrence theory, it is usual the study of the time elapsed until a fixed sequence x_0^k-1 appears in a given realization of process. This quantity is known as the hitting time and it is usually exponentially large in relation to the size of word. On the opposite, the first possible return time of a given word is defined as the minimum among all the hitting times of realizations that begins with the given word x_0^k-1. This quantity is tipically small that is of the order of the length of the sequence. In this work, we study the probability of the first possible return time given that the process begins of the target word. This quantity measures the intensity of that, once observed the target set, it can be observed in clusters. We show that, under certain conditions, the exponential decay rate of this probability converges to the entropy for all almost every word x_0^k-1 as k diverges. We also present an estimator of this probability for context trees and shows its consistency.

Considere um processo estocástico X_m em tempo discreto definido sobre o alfabeto finito A. Seja x_0^k-1 uma palavra fixa sobre A^k. No estudo das propriedades estatísticas na teoria de recorrência de Poincaré, é clássico o estudo do tempo decorrente até que a sequência fixa x_0^k-1 seja encontrada em uma realização do processo. Tipicamente, esta é uma quantidade exponencialmente grande com relação ao comprimento da palavra. Contrariamente, o primeiro tempo de retorno possível para uma sequência dada está definido como sendo o mínimo entre os tempos de entrada de todas as sequências que começam com a própria palavra e é uma quantidade tipicamente pequena, da ordem do tamanho da palavra. Neste trabalho estudamos o comportamento da probabilidade deste primeiro retorno possível de uma palavra x_0^k-1 dado que o processo começa com ela mesma. Esta quantidade mede a intensidade de que, uma vez observado um conjunto alvo, possam ser observados agrupamentos ou clusters. Provamos que, sob certas condições, a taxa de decaimento exponencial desta probabilidade converge para a entropia para quase toda a sequência quando k diverge. Apresentamos também um estimador desta probabilidade para árvores de contexto e mostramos sua consistência. / Considering a stochastic process X_m in a discrete defined time over a finite alphabet A and x_0^k-1 a fixed word over A^k. In the study of the statistical properties of the Poincaré recurrence theory, it is usual the study of the time elapsed until a fixed sequence x_0^k-1 appears in a given realization of process. This quantity is known as the hitting time and it is usually exponentially large in relation to the size of word. On the opposite, the first possible return time of a given word is defined as the minimum among all the hitting times of realizations that begins with the given word x_0^k-1. This quantity is tipically small that is of the order of the length of the sequence. In this work, we study the probability of the first possible return time given that the process begins of the target word. This quantity measures the intensity of that, once observed the target set, it can be observed in clusters. We show that, under certain conditions, the exponential decay rate of this probability converges to the entropy for all almost every word x_0^k-1 as k diverges. We also present an estimator of this probability for context trees and shows its consistency.

nÃo hà / O crescimento do uso e ocupaÃÃo dos solos em Ãreas urbanas e rurais exige conhecimento e tratamento estatÃstico de dados de intensidade, duraÃÃo e freqÃÃncia de chuvas intensas, importantes na previsÃo das vazÃes de pico utilizadas no dimensionamento de redes de drenagem. Os transtornos causados por alagamentos ou inundaÃÃes de Ãreas urbanas e rurais somente sÃo evitados com obras de drenagem bem dimensionadas, que resultam em custos de construÃÃo compatÃveis com os fatores de risco e adequada previsÃo da vazÃo de projeto. Estas vazÃes de projetos podem ser estimadas a partir da equaÃÃo IDF de chuva, que relaciona uma dada intensidade (I) de uma chuva intensa com a duraÃÃo (D) e a frequÃncia (F) ou tempo de retorno da mesma. Esta equaÃÃo à obtida para uma determinada localidade, geralmente por mÃtodo convencional, atravÃs de registros pluviogrÃficos de chuvas consideradas intensas para certas duraÃÃes e freqÃÃncias observadas. Entretanto, no Brasil nÃo existe registros pluviogrÃficos em quantidade suficiente, que abranja muitas Ãreas de importÃncia urbana, agrÃcola e industrial. No Estado do CearÃ, recentemente foram instalados 70 pluviÃgrafos digitais tipo bÃscula, monitorados pela FundaÃÃo Cearense de Meteorologia e Recursos HÃdricos (FUNCEME), porÃm com registros inferiores a quatro anos de observaÃÃo. Neste contexto, esta pesquisa objetivou desenvolver uma metodologia para obtenÃÃo da equaÃÃo IDF de chuvas intensas, em Ãreas desprovidas de pluviÃgrafos, utilizando dados obtidos de pluviÃmetros e tratados pelo mÃtodo das isozonas. Este trabalho foi desenvolvido em quatro etapas: (i) obtenÃÃo e anÃlise estatÃstica descritiva de sÃrie de precipitaÃÃes mÃximas anuais; (ii) obtenÃÃo do ajuste dos dados a uma funÃÃo distribuiÃÃo de probabilidade; (iii) geraÃÃo de intensidades de chuvas atravÃs do mÃtodo das isozonas; (iv) formulaÃÃo da equaÃÃo IDF de chuvas intensas. Esta metodologia foi aplicada aos municÃpios de Juazeiro do Norte, Barbalha e Crato, localizados na RegiÃo Metropolitana do Cariri cearense. Antes do desenvolvimento da metodologia foi feito a validaÃÃo do mÃtodo das isozonas em duas zonas climatologicamente distintas, comparando com intensidades obtidas de equaÃÃes IDF jà formuladas. A validaÃÃo mostrou, para os municÃpios de Fortaleza no Cearà e Miracema do Norte em Tocantins, que o mÃtodo das isozonas superestima as intensidades em aproximadamente 10%, podendo ser utilizado com certa seguranÃa. Ainda, a validaÃÃo do mÃtodo das isozonas foi verificada a partir de dados pluviogrÃficos da estaÃÃo climatolÃgica de Aiuaba â CE, localizada na isozona G. Os nÃmeros obtidos da estaÃÃo de Aiuaba mostram o bom desempenho da metodologia das isozonas, tendo importÃncia significativa para este trabalho, pois a Ãrea de estudo fica tambÃm na isozona G. AtravÃs da metodologia, para cada municÃpio estudado, o mÃtodo das isozonas transformou os dados bÃsicos de precipitaÃÃes diÃrias mÃximas anuais, obtidos dos registros pluviomÃtricos, em sÃries de intensidades mÃximas, duraÃÃes e tempos de retornos, que foram utilizados na formulaÃÃo das respectivas equaÃÃes IDF de chuvas intensas, para duraÃÃes de atà 240 min. As utilizaÃÃes destas equaÃÃes tornam a aplicaÃÃo do mÃtodo das isozonas mais prÃtica, eliminando a anÃlise de mapa e tabelas, reduzindo a possibilidade de erros de transcriÃÃo. / The increase in use of urban and rural land demands knowledge on statistically treated data on intensity, duration and frequency of heavy rainfall, allowing better forecasting of peak flow rates - used in the design of drainage networks. The damage caused by flooding of urban and rural areas are only avoided by well dimensioned drainage works, resulting in construction costs compatible with adequate treatment of the risk factors and with projected flow prediction. These projected flows can be obtained from the rainfall IDF equation, which links a given intensity (I) of heavy rain with its duration (D), and to its frequency (F) or with the time to pass until its return. This equation is obtained for a given location, usually by a conventional method, using rain gauge rainfall records considered intense for observed durations and frequencies. However, in Brazil there is insufficiency of rain gauge records, for many large and important urban, agricultural and industrial areas. Thus, 70 digital recording rain gauges were recently installed within the state of CearÃ, monitored by the FundaÃÃo Cearense de Meteorologia e Recursos HÃdricos (FUNCEME), but still with less than four years of recorded observation. In this context, this research was aimed at developing a methodology to obtain the equation of heavy rainfall IDF in areas devoid of recording rain gauges, using data obtained from conventional rain gauges, treated by the method of isozones. This study was conducted in four stages: (i) obtention and descriptive statistical analysis of a series of annual maximal precipitations, (ii) obtention of the data fit to a distribution function, (iii) generation of the rainfall intensities by the isozones method (iv) formulation of the heavy rainfall IDF equation. This methodology was applied to the municipalities of Juazeiro, Barbalha and Crato, located in the Cariri (CE) metropolitan area. Before the development of the methodology, it was performed the validation of the isozones method in two climatologically different areas, comparing it with intensities obtained from pre-existing IDF equations. The validation showed, for the cities of Fortaleza (in CearÃ) and Miracema do Norte (in Tocantins), that the isozones method overestimates the rainfall intensities in approximately 10%, and thus can be used with relative safety. Still, the validation of the isozones method was verified using rainfall data from the meteorological station of Aiuaba - EC, located in the isozone G. The numbers obtained from the Aiuaba station show the good performance of the isozones methodology, what is significantly important to this study, because the studied area is also in the isozone G. Through the methodology, for each city studied, the isozones method turned the basic data of annual daily maximal precipitation, in a series of intensities, durations and return times, which were used in the formulation of their heavy rainfall IDF equations, for durations of up to 240 min. These equations make the use of the isozones method more practical, eliminating the analysis of charts and tables, and thus reducing the probability of transcription errors.

Estudamos nesta dissertação o tempo de recorrência em sistemas dinâmicos, concentrando-nos na estatística do tempo de retorno. Calculamos numericamente a distribuição de tempo de retorno a uma região específica do espaço de fases de sistemas caóticos e comparamos com a distribuição binomial, deduzida para um processo aleatório. Os principais resultados obtidos foram: surgimento do efeito que denominamos memória de curto alcance, típico de sistemas determinísticos e associado à distribuição das órbitas periódicas instáveis; a distribuição de tempo de retorno caracteriza as principais propriedades temporais no caso de sistemas intermitentes. As conexões do tempo de retorno com regimes de transporte anômalo foram apresentadas, ressaltando suas limitações. O tempo de retorno foi utilizado ainda para analisar séries temporais, obtidas tanto de um modelo de mistura de um contaminante escalar passivo, como experimentalmente no plasma confinado magnéticamente. No primeiro caso constatamos que os retornos da série temporal assemelham-se às recorrências no espaço de fases do sistema dinâmico responsável pela mistura do contaminante: o mapa padrão com fase aleatória. Constatamos o surgimento de caudas de lei de potência na distribuição de tempo de retorno e calculamos sua dependência com o aumento da não linearidade e da aleatoriedade do sistema. Destacamos o efeito de múltiplas caudas de lei de potência, ausente no caso das distribuições obtidas no espaço de fases. Às séries obtidas em Tokamaks aplicamos o modelo de cascata log-normal para explicar sua função densidade de probabilidade. A distribuição de tempo de retorno destas séries mostrou estar diretamente relacionada com a correlação de curto e longo alcance presente na série. / We study the recurrence time in dynamical systems. The statistics of the recurrence time to a specific region of the phase space of chaotic dynamical systems were obtained numerically and compared with the binomial-like distribution, deduced for a random process. The main results are: the presence of the so called short time memory effect, typical for deterministic systems and related to the distribution of the unstable periodic orbits; the return time distribution captures the main temporal properties of intermittent systems. The possible connections of the recurrence time statistics to the anomalous transport were presented, with special attention to their limitations. The return time statistics was applied to analyze time series obtained from an Hamiltonian model and from magnetically confined plasma. In the first case we noticed that the recurrences of the series were similar to the recurrences obtained in the phase space of the Hamiltonian dynamical system: the standard map with a random phase. We analyze the dependence of the power-law tails of the distributions with the non-linearity and with the randomness of the system. One effect that appears only in the time series case is the multiple power law tails. We apply the log-normal cascade model to explain the probability density function of the series obtained in Tokamaks. The recurrence time statistics of the series is closely related to the short and long time correlation present on the series.

Estudamos nesta dissertação o tempo de recorrência em sistemas dinâmicos, concentrando-nos na estatística do tempo de retorno. Calculamos numericamente a distribuição de tempo de retorno a uma região específica do espaço de fases de sistemas caóticos e comparamos com a distribuição binomial, deduzida para um processo aleatório. Os principais resultados obtidos foram: surgimento do efeito que denominamos memória de curto alcance, típico de sistemas determinísticos e associado à distribuição das órbitas periódicas instáveis; a distribuição de tempo de retorno caracteriza as principais propriedades temporais no caso de sistemas intermitentes. As conexões do tempo de retorno com regimes de transporte anômalo foram apresentadas, ressaltando suas limitações. O tempo de retorno foi utilizado ainda para analisar séries temporais, obtidas tanto de um modelo de mistura de um contaminante escalar passivo, como experimentalmente no plasma confinado magnéticamente. No primeiro caso constatamos que os retornos da série temporal assemelham-se às recorrências no espaço de fases do sistema dinâmico responsável pela mistura do contaminante: o mapa padrão com fase aleatória. Constatamos o surgimento de caudas de lei de potência na distribuição de tempo de retorno e calculamos sua dependência com o aumento da não linearidade e da aleatoriedade do sistema. Destacamos o efeito de múltiplas caudas de lei de potência, ausente no caso das distribuições obtidas no espaço de fases. Às séries obtidas em Tokamaks aplicamos o modelo de cascata log-normal para explicar sua função densidade de probabilidade. A distribuição de tempo de retorno destas séries mostrou estar diretamente relacionada com a correlação de curto e longo alcance presente na série. / We study the recurrence time in dynamical systems. The statistics of the recurrence time to a specific region of the phase space of chaotic dynamical systems were obtained numerically and compared with the binomial-like distribution, deduced for a random process. The main results are: the presence of the so called short time memory effect, typical for deterministic systems and related to the distribution of the unstable periodic orbits; the return time distribution captures the main temporal properties of intermittent systems. The possible connections of the recurrence time statistics to the anomalous transport were presented, with special attention to their limitations. The return time statistics was applied to analyze time series obtained from an Hamiltonian model and from magnetically confined plasma. In the first case we noticed that the recurrences of the series were similar to the recurrences obtained in the phase space of the Hamiltonian dynamical system: the standard map with a random phase. We analyze the dependence of the power-law tails of the distributions with the non-linearity and with the randomness of the system. One effect that appears only in the time series case is the multiple power law tails. We apply the log-normal cascade model to explain the probability density function of the series obtained in Tokamaks. The recurrence time statistics of the series is closely related to the short and long time correlation present on the series.

Forskning visar att den globala uppvärmningen kan orsaka förändrade nederbördsmönster på många platser runt om i världen. Ett antal platser kan förvänta sig en ökad nederbördsmängd under delar av året, medan andra platser kan förvänta sig längre perioder av torka. SMHI presenterar prognoser över hur nederbördsmönster kan ändras i Sverige till följd av klimatförändringar. I Jönköpings län förväntas den mest kraftfulla nederbörden att öka, och även årsnederbörden. Då översvämningar kan leda till försämrad framkomlighet, skador eller i värsta fall hot mot hälsa och liv har EU tagit fram ett översvämningsdirektiv. Direktivet ska assistera unionens medlemmar att förebygga och hantera problematik som uppstår i samband med översvämningar. I linje med EU:s översvämningsdirektiv har Sverige arbetat fram olika strategier för att kunna förebygga problem som uppstår i samband med översvämningar. I första hand ska kartläggningar av områden utföras för att lokalisera och identifiera områden som är utsatta för översvämningsrisker. Alla områden som är utsatta för översvämningsrisker ska sedan ta fram riskhanteringsplaner där olika strategier och förebyggande åtgärder presenteras. WSP har utfört skyfallskarteringar åt bland annat Tranås kommun i Jönköpings län. I denna rapport har ett område i Tranås valts ut för att studera hur stor effekt olika åtgärder kan ha vid en översvämning till följd av ett simulerat 100-årsregn. WSP:s skyfallskartering har använts vid val av område och Tranås kommun har även konsulterats. Åtgärder har implementerats vid området i en modell som sedan använts för att utföra simuleringar. Området som valts ut består av en parkeringsyta med en intilliggande grönyta. Vid platsen finns en befintlig dam som är i mediokert skick. Åtgärderna består av olika dimensioner av torrdammar och av ett svackdike. Modifiering av terrängfiler har utförts med hjälp av det webbaserade mjukvaruprogrammet Scalgo. Bearbetning av data har utförts i GIS och i MIKE11. Simuleringarna har gett svar på vilka effekter de olika åtgärderna kan ha. Resultaten visar att för att kunna använda parkeringsytan krävs att en relativt stor torrdamm utformas vid den intilliggande grönytan. En magasinsvolym på 4970 kubikmeter möjliggör att 90 procent att parkeringsytan går att använda obehindrat. Resultatet bygger på att en utloppsledning kan ha dimensionen 300 mm. I dagsläget finns en utloppsledning med dimensionen 210 mm. En utredning kan resultera i vetskapen om ledningens dimension går att utöka till 300 mm. Resultaten visar att torrdammar med svackdiken har en potential att förebygga översvämningar i tätbebyggda, urbana områden. Effekterna av de olika åtgärderna har presenterats i en tabell. / Global warming is causing changes in precipitation patterns world over. Many places can expect to face longer periods of drought, while others may be facing periods with increased amounts of precipitation. Europe has created a Flooding Directive in order to prevent consequences of flooding. In Sweden a work of mapping areas with risk of flooding has been a first step to fulfill the Directive. Secondly, all areas that are affected shall create risk plans where their different strategies and work plans are being presented. WSP have mapped areas in Tranås in Jönköping, Sweden, in order to find out which areas could be the most vulnerable to flooding after a heavy downpour. The mapping is being performed with the software Mike+ that is developed by DHI, and the rain event that has been used is a 100-year downpour with an added climate factor. In this report one of the mapped areas in Tranås, has been selected in order to present proposed measures. The measures consist of dry ponds and ditches and are being designed with different dimensions. Simulations have been performed in Mike+ in order to study the different measure‘s effects after a 100-year storm event. The results show that the implemented dry pond needs to be rather big to show good effect. The best effect was shown with the dry pond of 4970 square meters and with an outlet line that had a dimension of 300 mm, which was 80 mm larger than the original. The result showed that about 90 percent of the parking lot could be used after the implementation. Furthermore, the results showed that with a pond measuring 4142 cubic meters, about 50 percent of the parking lot could be used. Overall, dry ponds showed potential of working as a measure against flooding in the area. Urban areas that are densely built could benefit from having dry ponds and ditches implemented. Their flexibility is valuable since they can be used for different purposes during dry periods, which can be useful in areas that are densely built.

Dans cette thèse, nous étudions les propriétés quantitatives de récurrence de certains systèmes dynamiques préservant une mesure infinie. Nous nous intéressons au premier temps de retour des orbites d'un système dynamique dans un petit voisinage de leurs points de départ. Tout d'abord, nous commençons par considérer un modèle jouet probabilistique pour éclairer la stratégie de nos preuves. On s'intéresse particulièrement au cas où la mesure est infinie, plus précisément, nous considérons les Z -extensions des sous-shift de type fini. Nous étudions le comportement asymptotique du premier temps de retour au voisinage de l'origine, et nous établissons des résultats de type de convergence presque partout, et aussi de convergence en loi par rapport à toute mesure de probabilité absolument continue par rapport à la mesure infinie. Dans ce travail, nous nous également intéressons à d'autres systèmes dynamiques. Nous considérons un flot Axiome A(gt)t sur une variété riemannienne M munie d'une mesure σ -finie μ. Nous supposerons que la mesure μ est une mesure d'équilibre pour (gt)t. Afin d'établir nos résultats, nous introduisons des notions de dynamique hyperbolique. En particulier, nous considérons la section de Markov qui a été introduite par Bowen et Ratner. / In this thesis, we study the quantitative recurrence properties of some dynamical systems preserving an infinite measure. We are interested in the first return time of the orbits of a dynamical system into a small neighborhood of their starting points. First, we start by considering a toy probabilistic model to clarify the strategy of our proofs. Our interest is when the measure is indeed infinite, more precisely we consider the Z-extensions of subshifts of finite type. We study the asymptotic behavior of the first return time near the origin, and we establish results of an almost everywhere convergence kind, and a convergence in distribution with respect to any probability measure absolutely continuous with respect to the infinite measure. In this work, we are also interested in another dynamicals systems. We consider an Axiom A flow (gt)t on a Riemannian manifold M endowed with a σ-finite measure μ. We will assume that the measure μ is an equilibrium measure for (gt)t. In order to establish our results, we introduce notions from hyperbolic dynamics. In particular, we consider the Markov section which was constructed by Bowen and Ratner.

This diploma thesis solves the evaluation of the economic efficiency of the investment - Municipal Library in Třinec. The aim of the work was to verify the effectiveness of this project using the CBA analysis. The theoretical part generally describes concepts such as investments, public contracts, public projects, the life stage of the project and, above all, the evaluation of the effectiveness of investment projects. The practical part deals with the effectiveness of the investment from the operational data provided by the library, which is compared with the planned variant.

Syfte: Syftet med studien är att undersöka hur e-handelsföretags returpolicys, mer specifikt returkomponenterna returtid, returkostnad och returansträngning, kan påverka konsumenters upplevda risk vid köp av kläder online. Studien utgår från tre hypoteser, vilka är: “Längre returtider resulterar i en lägre upplevd risk hos konsumenter vid köp av kläder online”, “Lägre returkostnader leder till lägre upplevd risk hos konsumenter vid köp av kläder online” samt “Lägre returansträngning medför att konsumenter upplever en lägre risk vid köp av kläder online”. Metod: Tvärsnittsstudien omfattades av en enkätundersökning där deltagarnas åsikter och inställningar ställdes i centrum. 109 användbara svar genererades utifrån ett bekvämlighetsurval. Svaren analyserades genom en multipel regressionsanalys för att undersöka respektive returkomponents påverkan på konsumentens upplevda risk vid köp av kläder online. Resultat: Resultatet indikerade på att ökad returtid respektive en minskad returkostnad bidrar till att reducera den upplevda risken för konsumenten vid köp av kläder online. Vad gäller returansträngning fann studien inget stöd för att en lägre ansträngning genererade en lägre upplevd risk för konsumenten vid köp av kläder online. Däremot saknades signifikans för variablerna returtid och returansträngning vilket gör det svårt att utesluta slumpfaktorn i dessa hypoteser. Det gick även att observera att returkomponenten returkostnad påverkade den upplevda risken i störst omfattning. Originalitet/Värde: I en ständigt växande e-handelsmarknad, speciellt inom modebranschen, är konkurrensen ett faktum vilket gör att företagen alltmer konkurrerar i andra områden än de produkter de säljer. Det finns en problematik kring hur returpolicys ska utformas på bästa sätt. Då upplevd risk har en betydande effekt på köp- och returintentionen av kläder köpta av konsumenter online är det ett relevant och viktigt ämne att studera. Studien kan bidra med insikt i vilka returkomponenter som värderas och uppskattas mest av konsumenter i en returpolicy, något som kan påverka konsumenters inställning till köpet. Med grund i studien och dess resultat konstateras att företag som säljer kläder online bör arbeta med att utforma sina returpolicys på ett sätt som minskar konsumenternas upplevda risk i största möjliga utsträckning. Vid situationer där företag behöver strama åt sina returpolicys kan denna studie hjälpa till med att ge stöd för vilken returkomponent som värderas minst av konsumenter. / Purpose: The aim of the study is to investigate how e-commerce companies’ return policies, more specifically the return components return time, return cost and return effort, can affect consumers' perceived risk while purchasing clothes online. The study revolves around three different hypotheses, which are: “Longer return times result in a lower perceived risk for consumers when buying clothes online”, “Lower return costs lead to a lower perceived risk among consumers when buying clothes online” as well as “Lower return efforts lead to consumers experiencing a lower risk when buying clothes online” Methodology: The cross-sectional study consists of a questionnaire survey in which the participants' opinions and attitudes were at the center. 109 useful responses were generated based on a convenience sample. The responses were analyzed through a multiple regression analysis, in order to examine the impact of each return component on the consumers' perceived risk while purchasing clothes online. Findings: The result indicated that an increased return time as well as a reduced return cost contribute to reducing the perceived risk for the consumer when purchasing clothes online. Regarding return effort, the study found no support that a lower return effort generated a lower perceived risk for the consumer when buying clothes online. However, the findings regarding return time and return effort were lacking significance, which means that there is a risk of the findings being coincidental. It was also possible to observe that the return component return cost affected the consumers’ perceived risk to the greatest extent. Originality / Value: In an ever-growing e-commerce market, especially in the fashion industry, competition is a fact which results in companies increasingly cofmpeting in other areas than the products they sell. There is a problem with how return policies should be designed in the best way. As perceived risk has an effect on the consumers’ intention to buy and return clothes purchased online, it is a relevant and important topic to study. The study can provide insight into which return components consumers value and appreciate most in a return policy, something that can affect consumers' attitudes towards the purchase. Based on the study and its results, it is stated that companies that sell clothes online should work on designing their return policies in a way that reduces consumers' perceived risk to the greatest possible extent. In situations where companies have to tighten their return policies, this study can help provide support for which return component is least valued by consumers. This thesis is written in Swedish.