O estudo de séries temporais para a geração de conhecimento é uma área que vem crescendo em importância e complexidade ao longo da última década, à medida que a quantidade de dados armazenados cresce exponencialmente. Considerando este cenário, novas técnicas de mineração de dados têm sido constantemente desenvolvidas para lidar com esta situação. Neste trabalho é proposto o estudo de séries temporais baseado em suas características topológicas, observadas em uma rede complexa gerada com os dados da série temporal. Especificamente, o objetivo do modelo proposto é criar um algoritmo de detecção de tendências para séries temporais estocásticas baseado em detecção de comunidades e caminhadas nesta mesma rede. O modelo proposto apresenta algumas vantagens em relação à métodos tradicionais, como o número adaptativo de classes, com força mensurável, e uma melhor absorção de ruídos. Resultados experimentais em bases artificiais e reais mostram que o método proposto é capaz de classificar as séries temporais em padrões locais e globais, melhorando a previsibilidade das séries ao se utilizar métodos de aprendizado de máquina para a previsão das classes / Extracting knowledge from time series analysis has been growing in importance and complexity over the last decade as the amount of stored data has increased exponentially. Considering this scenario, new data mining techniques have continuously developed to deal with such a situation. In this work, we propose to study time series based on its topological characteristics, observed on a complex network generated from the time series data. Specifically, the aim of the proposed model is to create a trend detection algorithm for stochastic time series based on community detection and network metrics. The proposed model presents some advantages over traditional time series analysis, such as adaptive number of classes with measurable strength and better noise absorption. Experimental results on artificial and real datasets shows that the proposed method is able to classify the time series into local and global patterns, improving the predictability of the series when using machine-learning methods

Submitted by Danilo José Brandao Vitor Silva null (danilojbrandao@yahoo.com.br) on 2017-02-03T15:40:19Z No. of bitstreams: 1 TESE DANILO BRANDÃO.pdf: 1732370 bytes, checksum: 2cb226321b60bf10a72d6f5d86091dfb (MD5) / Approved for entry into archive by LUIZA DE MENEZES ROMANETTO (luizamenezes@reitoria.unesp.br) on 2017-02-07T11:31:33Z (GMT) No. of bitstreams: 1 silva_djbv_dr_guara.pdf: 1732370 bytes, checksum: 2cb226321b60bf10a72d6f5d86091dfb (MD5) / Made available in DSpace on 2017-02-07T11:31:33Z (GMT). No. of bitstreams: 1 silva_djbv_dr_guara.pdf: 1732370 bytes, checksum: 2cb226321b60bf10a72d6f5d86091dfb (MD5) Previous issue date: 2017-01-19 / Este trabalho analisou os parâmetros hidroclimatológicos, buscando encontrar o padrão de precipitação em uma abordagem espaço-temporal para a porção paulista da bacia do rio Paraíba do Sul, empregando o interpolador geoestatístico krigagem. Além disso, verificou a existência de tendência de precipitação e de vazão em séries históricas nesta mesma bacia, utilizando o teste estatístico de Mann-Kendall. Com relação a espacialização da precipitação, foi constatada uma significativa variação espacial para os períodos analisados (verão, outono, inverno e primavera). No verão, período de recarga dos reservatórios, foram registrados os menores índices de precipitação na área de contribuição das sub-bacias que abastecem os reservatórios do Paraibuna e do Jaguari, implicando em uma possível redução na vazão afluente. Não se constatou tendência de precipitação estatisticamente significativa em escala de tempo interanual para os postos pluviométricos, entretanto, foi identificada tendência de diminuição para o outono. Para os postos fluviométricos das sub-bacias do rio Paraibuna, do rio Paraitinga e do rio Jaguari-Buquira, em um período interanual, houve tendência de aumento de vazão para séries históricas de 1930 a 1970 e de diminuição para séries de 1930 a 2015. As estações de medição de vazão utilizadas foram aquelas localizadas a montante dos reservatórios, não estando submetidas assim ao efeito de regularização de vazão. A tendência de diminuição de vazão encontrada para a região pode levar a uma alteração na disponibilidade hídrica e a um possível impacto no abastecimento e na geração de energia. / This research analyzed the hydroclimatological parameters, seeking to find the precipitation pattern in a space-time approach for the São Paulo portion of the Paraíba do Sul river basin, using the geostatistical interpolator kriging. In addition, it verified the existence of precipitation and streamflow trends in historical series for the same basin, using the statistical test of Mann-Kendall. Regarding the precipitation pattern, a significant spatial variability was observed for the periods analyzed (summer, autumn, winter and spring). In the summer, reservoir recharge period, the lowest precipitation rates were recorded in the sub-basin contribution area that supply the Paraibuna and Jaguari reservoirs, implying a possible reduction in the inflow. There was no statistically significant precipitation trend in the interannual time scale for the pluviometric stations, however, a decrease trend was identified for the autumn. For fluviometric stations in the sub-basins of the Paraibuna, Paraitinga and Jaguari-Buquira rivers, in the interannual time scale, there was a trend towards an increase in streamflow for historical series from 1930 to 1970 and decrease for series from 1930 to 2015. The streamflow stations used were those located upstream of the reservoirs and were not subject to the effect of flow regulation. The downward trend in streamflow found in the region can lead to a change in water availability and a possible impact on the supply and energy generation.

A deriva litorânea é estudada no litoral sul do Espírito Santo através de diferentes técnicas: distribuição de Minerais Pesados (MP); modelo conceitual Análise de Transport de Sedimento (Sediment Trend Analysis - STA), através da variação de parâmetros texturais do sedimento; dados de clima de onda entre 1997 e 2010 do modelo global WaveWatch III propagados para a zona costeira através do modelo Delft-3D, fornecendo parâmetros necessários para as equações de deriva; e imagens de satélite e fotografias aéreas para avaliar a variação espaço-temporal da morfologia costeira. O modelo STA mostrou-se confiável apenas nos dois setores ao norte, enquanto nos demais os resultados foram mascarados por altos teores de MP e/ou carbonato e a entrada de outras fácies no sistema; as equações representaram bem a variação da morfologia costeira na área de estudos. De maneira geral, os setores ao sul apresentaram transporte longitudinal para sul, e os setores ao norte, para norte, fato relacionado à batimetria local. A orientação da linha de costa auxiliou na compreensão dos fenômenos erosivos, pois em sua maioria estão orientados no sentido NEN, formando 45º com as ondas dos setores de ESE e SE. / Longshore drift is studied in the southern portion of Espírito Santo through different techniques: Heavy Minerals (HM) distribution; the conceptual model Sediment Trend Analysis (STA) by sediment textural parameters fluctuation; wave climate data between 1997 and 2010 from WaveWatch III propagated to the coastal zone through Delft-3D model, providing the required data for longshore drift equations; and satellite images and aerial photograph to survey time-space coastline variations. The STA model appears to be trustworthy only within the northern sectors, as high HM and/or carbonate content and different facies input into the system may mask the results. Broadly speaking, the southern sectors showed longshore transport southward, while northern sectors showed northward longshore transport, which is related to local bathymetry. Shoreline azimuth helped understanding the location of erosive zones, since most of them are turned NEN, 45o with ESE and SE wave sectors.

Detection of structural changes in time series is a topic with increasing pop- ularity among econometricians over the last decades. The main aim of this thesis was to review and compare the classical and modern econometric meth- ods of structural change detection and unit root testing. A recent method for testing a one-time break in at most linear trend function of a series without prior knowledge about the stationary or unit root nature of the error compo- nent proposed by Perron and Yabu (2009b) was studied. Subsequently, it was combined with the unit root test that allows for a break in trend proposed by Kim and Perron (2009) to examine the nature of the error component. All the methods for change detection and unit root testing were compared in a Monte Carlo simulation study that indicated significant improvement in power of the Perron-Yabu and Kim-Perron tests against most alternatives compared to the classical methods. However, all tests demonstrated poor performance in case of a quadratic trend function. Finally, the tests were employed in a practical ex- ample to examine the properties of the quarterly GDP time series of the Czech Republic. 1

Aim: By studying the evolution of the market for package holidays over time we hope to get an understanding of the trends that have characterized the market. By doing so, we can get an idea of where the market is moving towards the future. This knowledge is essential for market participants to establish appropriate strategies that follow market trends. Method: We have used qualitative interviews with people involved in the charter industry. Result & Conclusions: The market is characterized by an increase in Internet orientation, where the travel directory and the physical travel agents are becoming less important. The understanding of climate impacts of travel has increased. More people of the third-generation buy package travels and countries in North Africa and Asia are growing in popularity. We believe the market in the coming years will be characterized by lower prices and lower quality. Travel directories will be replaced by promotional flyers. Marketing will be aimed at the third generation. It will be difficult for physical retailers to retain customers. Interest in countries such as Vietnam, Malaysia and Egypt will increase. Suggestions for future research: “A study of the physical dealer prospects” and “Studies on the consumers’ willingness to pay for environmentally friendly charter” would be interesting.   Contribution of the thesis: An increased understanding of the trends in the charter industry.   Key words: Trend, charter companies, market, marketing, consumers, destinations, resorts, travel directory, Internet, development, future. / Syfte: Genom att studera utvecklingen på marknaden för charterresor över tid hoppas vi få förståelse för vilka trender som har präglat marknaden för charterresor och vilka tendenser vi kan se att utvecklingen rör sig mot. Denna kunskap är nödvändig för att charterbolagen ska kunna upprätta relevanta strategier som följer marknadens utveckling. Metod: Vi har använt oss av kvalitativa intervjuer med personer verksamma inom charterbranschen. Resultat och slutsats: Marknaden karaktäriseras av en ökad internetorientering, där resekatalogen och de fysiska resebyråerna blir allt mindre betydelsefulla. Även kunskapen om resandets klimateffekter har ökat. Allt fler personer av den äldre generationen köper charterresor och länder i Nordafrika och Asien växer i popularitet. Vi bedömer att marknaden de kommande åren kommer att kännetecknas av lägre priser och sämre kvalitet. Resekatalogen ersätts av reklamblad. Marknadsföring kommer att riktas till den äldre generationen. Det blir svårt för fysiska återförsäljare att behålla kunder. Intresset för länder som Vietnam, Malaysia och Egypten kommer att öka.   Förslag till fortsatt forskning: En kartläggning av de fysiska återförsäljarnas framtidsutsikter samt studier om betalningsviljan för miljövänliga charterresor vore intressant Uppsatsens bidrag: Ökad förståelse för utvecklingen i charterbranschen.   Nyckelord: Trend, charterbolag, marknad, marknadsföring, konsument, destination, resmål, resekatalog, internet, utveckling, framtid.

Logistikens vikt för företag blir allt större. En mängd faktorer i företagens omvärld tvingar företag att följa med i utvecklingen för att behålla sin konkurrenskraft. Den ständiga förändringen som sker inom logistikområdet präglas av en mängd tidstypiska trender. Företag med en betydande logistik tvingas därför ta ställning till dessa trender, som på olika sätt sätter avtryck i organisationer då de väljer att implementera tillhörande arbetssätt och metoder från trenderna. Studien undersöker hur företag påverkas av logistiska trender. Detta sker genom att beskriva hur de ser på utvecklingen inom logistik och de trender som utvecklas inom området, samt att se till hur trenderna har satt spår i företagens verksamheter. Supply Chain Management spelar en central roll för uppsatsen, då den statuerar exempel på en logistisk trend, tillsammans med Lean Production och Just-In-Time. Studien är genomförd genom tio semistrukturerade intervjuer hos företag som ser logistik som viktig för deras verksamhet, antingen ur en kostnadssynpunkt eller ur en konkurrenssynpunkt. Den intervjuade representanten för respektive företag är antingen ansvarig för verksamhetens logistik eller en logistikfunktion. Studiens slutsatser är att de företag som intervjuats alla kommer i kontakt med trender inom logistik. Sättet en trend förs in i organisationen sker främst genom nyanställningar av personal med teoretisk eller praktisk bakgrund från andra verksamheter, eller genom att inspireras av andra företag. För att det ska vara aktuellt för företag att införa delar av en trend som Just-In-Time, Lean Production eller Supply Chain Management krävs det att dessa genererar en effektivisering av verksamheten, främst genom ekonomiska besparingar. Supply Chain Management har gjort avtryck hos studiens företag i det avseendet att det finns ett stort fokus på att utveckla samarbetet med sina leverantörer samt att det finns en strävan efter att öka företagets kundfokus och därmed sätta kunden i centrum för verksamheten. / The importance of logistics to companies is growing. A number of factors in business environment forces companies to keep up with changes to remain competitive. The constant changes that occur in the logistics area are characterized by a variety of contemporary trends. Companies with a major logistics are therefore forced to consider these trends, which in various ways puts a mark on the organizations where they choose to implement the related procedures and methods of trends. This study examines how businesses are affected by logistical trends. This is done by describing how they view the developments in logistics and the trends that develop in the area, and to see how trends have left traces in the activities of businesses. Supply Chain Management plays a central role for the study, as it sets an example of a logistic trend, along with Lean Production and Just-In-Time. The study was conducted by ten semi-structured interviews with companies who see logistics as important to their business, either from a cost perspective or from a competitive standpoint. The interviewed representative of the respective companies are either responsible for the businesses logistics or a logistics function. The conclusions of the study are that the companies interviewed all come in contact with trends in logistics. The way a trend is brought into the organization is done primarily through the hiring of staff with practical or theoretical background from other activities, or to be inspired by other companies. For it to be appropriate for companies to implement the elements of a trend such as Just-In-Time, Lean Production or Supply Chain Management, it is required that they generate corporate efficiency, mainly through financial savings. Supply Chain Management has made an important impact in the study’s companies in the sense that there is a major focus on developing cooperation with its suppliers and that there is a desire to increase its customer focus and thus put the customer at the center of activity.

Sedan millenniumskiftet förändrades modevärlden radikalt av fler säsonger, ökad mångfald av trender samtidigt som trendernas livslängds blev allt kortare. Hos inköparna på de stora klädbolagen ställs det idag ett allt större krav att veta vad som kommer att sälja i framtiden. Idag måste besluten tas snabbt så att kläderna hinner producera och levereras i tid innan modet hinner ändra sig. Som ett hjälpmedel för inköpare och designers om i vilken riktning modet kommer att gå finns trender som fungerar som ett avgränsat modeuttryck för att förmedla och tolka en viss stil och därmed ge en tydlig bild om vad som bör produceras och efterfrågas. Men eftersom internet har medfört en ökad acceleration av trender kan det upplevas svårt att veta vad som kommer sälja i framtiden och det är här ser vi trendbyråernas roll. Trendbyråer har en koordinerande roll på marknaden där de ger företag vägledning om förändringar i modet. Genom trendanalyser får företag en överblick hur riktningen i modet kommer att gå, förstå vilka globala influenser som påverkar modet och en reflektion kring hur allt återspeglas till modet. Vad gäller tidigare forskning av begreppen trender och trendbyråer vid beslutsfattande, har ingen omfattande undersökning tidigare gjorts vilket har medfört att vår uppsats känns värdefull för en mer förståelse kring ämnet. Avsikten med vår uppsats är att undersöka och analysera hur trender och trendanalyser har för betydelse vid beslutsfattande inom inköpsprocessens tidiga stadium hos svenska modeföretag. Studien har efter insamling av teori inom områdena beslut, trender och trendanalyser genomförts med en kvalitativ grund där insamling av det empiriska materialet har utgått från semistrukturerade intervjuer. Intervjuer har främst genomförts med respondenter som innehar en roll inom inköp på svenska modeföretag som ingår i beteckningen SME (Small medium Enterprises). Dessa företag är Ellos, Lindex, Gina Tricot och MQ. För att öka förståelsen kring trender och trendanalyser har intervjuer även genomförts med en trendanalytiker från Svenska Moderådet och en författare bakom boken Trendmakarna. Insamlad data har därefter analyserats och ställts mot teorier angående beslut, trender och trendanalyser i syfte att kunna uttala oss om hur inköpare ser på trender och trendbyråers betydelse vid beslutsfattande av inköp. De resultat som empirin bestått av ger indikationer på att trender har en betydande roll men att det både kan öka och minska på osäkerheter samt risker. Inköparna anser att det finns många risker med trendbaserade produkter men att trendanalyser vid beslutsfattande kan användas för att förstå riktningen modet går i för att därefter kunna anpassa inköpen genom volym, budget, material och leverantörer. Därmed kan de vara ett hjälpmedel för att minska på både ett besluts risk och osäkerhet, vilket bekräftar de teorier som tagits upp. Dock kan vi uttala oss om att beslutsfattande inom inköp inte bara baseras på trendanalyser utan andra faktorer och variabler spelar också in.

The forecast of the waste production is an important information for planning in waste management. The historical data often consists of short time series, therefore traditional prognostic approaches fail. The mathematical model for forecasting of future waste production based on spatially distributed data with hierarchically structure is suggested in this thesis. The approach is based on principles of regression analysis with final balance to ensure the compliance of aggregated data values. The selection of the regression function is a part of mathematical model for high-quality description of data trend. In addition, outlier values are cleared, which occur abundantly in the database. The emphasis is on decomposition of extensive model into subtasks, which lead to a simpler implementation. The output of this thesis is tool tested within case study on municipal waste production data in the Czech Republic.

This diploma thesis is focused on the general characteristics of the technical analysis, definitions of the information sources and technical analyses. Main attention is given to the particular indicators of the technical analysis. Practical part of this thesis includes applying of the defined indicators in the evaluation of the money market situation.

Determination sex is one of the most important tasks in the identification of skeletal remains. Previous studies have shown that population differences in size and shape of femora can affect correct sex allocation. This thesis tested the discriminat functions generated from different populations and confirmed population specificity of the femur discriminant functions. Two samples of the identified adult femora for this project was used. First sample originates from the Czech population living in the 1st 20th century (78 males and 81 females) and second sample originates from the Czech population living in the 2nd 20th century (67 males and 35 females). Both samples also help us estimate the influence of secular trend of the femur'variables to the correct sex classification by discriminat functions calculated in the Czech population. The results showed the existence of the secular trend in femora size dimensions. Only the dimensions of the femoral head (vertical head diameter-F18, transverse head diameter-F19 and maximum diameter of the femoral head-MPH) and sagittal subtrochanteric diameter (F10) not chandged significantly during the 20th century. The secular trend has negatively influenced the correct sex classification, always for one subpopulation. Only the dimensions in which the secular trend...