Submitted by Luciclea Silva (luci@ufpa.br) on 2018-11-20T16:16:30Z No. of bitstreams: 2 license_rdf: 9 bytes, checksum: 42dd12a06de379d3ffa39b67dc9c7aff (MD5) Tese_Modelohibridobaseado.pdf: 2923843 bytes, checksum: 6646b898a2999050d56c3e291110b46d (MD5) / Approved for entry into archive by Luciclea Silva (luci@ufpa.br) on 2018-11-20T16:16:59Z (GMT) No. of bitstreams: 2 license_rdf: 9 bytes, checksum: 42dd12a06de379d3ffa39b67dc9c7aff (MD5) Tese_Modelohibridobaseado.pdf: 2923843 bytes, checksum: 6646b898a2999050d56c3e291110b46d (MD5) / Made available in DSpace on 2018-11-20T16:16:59Z (GMT). No. of bitstreams: 2 license_rdf: 9 bytes, checksum: 42dd12a06de379d3ffa39b67dc9c7aff (MD5) Tese_Modelohibridobaseado.pdf: 2923843 bytes, checksum: 6646b898a2999050d56c3e291110b46d (MD5) Previous issue date: 2018-08-30 / FAPEAM - Fundação de Amparo à Pesquisa do Estado do Amazonas / A geração de energia elétrica através de turbinas eólicas é uma das alternativas praticamente inesgotáveis de geração de energia elétrica. Ela é considerada uma fonte de energia limpa, porém ainda necessita de muita pesquisa para desenvolvimento de ciência e tecnologias que assegurem uma uniformidade na geração, propiciando uma maior participação desta fonte na matriz energética tanto no Brasil quanto no mundo, pois o vento apresenta bruscas variações na velocidade, densidade e em outras variáveis importantes. Nos sistemas elétricos de base eólica, cada horizonte de previsão é aplicado em um determinado segmento específico, previsão de minutos, horas, semanas, meses e anos futuros do comportamento do vento, desta forma pode-se avaliar a disponibilidade de energia para o próximo período, uma informação relevante no despacho das unidades geradoras e no controle do sistema elétrico. Esta tese teve como proposta, desenvolver modelos de previsão a ultra curto, curto, médio e longo prazo da velocidade do vento, baseado em técnicas de inteligência computacional, usando modelos de Redes Neurais Artificiais, SARIMA e modelos híbridos e prever a capacidade da geração de potência para cada horizonte. Para aplicação da metodologia utilizou-se as variáveis meteorológicas do banco de dados do sistema de organização nacional de dados ambientais SONDA, estação de Petrolina, do período de 01 de janeiro de 2004 à 31 de março de 2017. O desempenho dos modelos foi comparado com 5, 10 e 20 passos para frente, considerando minutos, horas, dias, semanas, meses e anos como horizonte de previsão. O modelo hibrido obteve melhor resposta na previsões dentre as quais destacou-se o horizonte de horas. / The electric power generation through wind turbines is one of the practically inexhaustible alternatives sources of electric power. It is considered a source of clean energy, but still requires a lot of research to develop science and technologies that ensure uniformity in generation, providing a greater participation of this source in the energy matrix in Brazil as in the world, because the wind presents abrupt variations speed, density, and other important variables. In wind-based electrical systems, each forecast horizon is applied to a specific segment, forecast of minutes, hours, weeks, months, and future years of wind behavior, in order to evaluate the availability of energy for the next period, relevant information in the dispatch of the generating units and in the control of the electric system. This thesis aimed to develop ultra-short, short, medium and long-term prediction models of wind speed, based on computational intelligence techniques, using Artificial Neural Networks, SARIMA models and hybrid models and to predict the generation capacity of power for each horizon. For the application of the methodology, the meteorological variables of the database of the national environmental data system SONDA, Petrolina station, were used for the period from January 1st, 2004 to March 31st, 2017. The performance of the models was compared with 5, 10 and 20 steps forward, considering minutes, hours, days, weeks, months and years as the forecast horizon. The hybrid model obtained better response in the forecasts, among which the hour horizon was highlighted.

Ghana faces a macroeconomic problem of inflation for a long period of time. The problem in somehow slows the economic growth in this country. As we all know, inflation is one of the major economic challenges facing most countries in the world especially those in African including Ghana. Therefore, forecasting inflation rates in Ghana becomes very important for its government to design economic strategies or effective monetary policies to combat any unexpected high inflation in this country. This paper studies seasonal autoregressive integrated moving average model to forecast inflation rates in Ghana. Using monthly inflation data from July 1991 to December 2009, we find that ARIMA (1,1,1)(0,0,1)12 can represent the data behavior of inflation rate in Ghana well. Based on the selected model, we forecast seven (7) months inflation rates of Ghana outside the sample period (i.e. from January 2010 to July 2010). The observed inflation rate from January to April which was published by Ghana Statistical Service Department fall within the 95% confidence interval obtained from the designed model. The forecasted results show a decreasing pattern and a turning point of Ghana inflation in the month of July.

Submitted by Julio Cesar de Azevedo Vieira (julio_vieira@globo.com) on 2018-06-16T14:57:18Z No. of bitstreams: 1 dissertacao_JulioCesarVieira.pdf: 1988173 bytes, checksum: 55cb349d2840d5de748cbd814f155bb9 (MD5) / Rejected by Marcia Bacha (marcia.bacha@fgv.br), reason: O aluno irá submeter com o novo PDF on 2018-06-19T14:38:11Z (GMT) / Submitted by Julio Cesar de Azevedo Vieira (julio_vieira@globo.com) on 2018-06-26T21:10:08Z No. of bitstreams: 1 dissertacao_JulioCesarVieira.pdf: 1801751 bytes, checksum: 382cab03be50d392c166a61e21222c05 (MD5) / Approved for entry into archive by Janete de Oliveira Feitosa (janete.feitosa@fgv.br) on 2018-07-05T13:19:09Z (GMT) No. of bitstreams: 1 dissertacao_JulioCesarVieira.pdf: 1801751 bytes, checksum: 382cab03be50d392c166a61e21222c05 (MD5) / Made available in DSpace on 2018-07-16T19:25:05Z (GMT). No. of bitstreams: 1 dissertacao_JulioCesarVieira.pdf: 1801751 bytes, checksum: 382cab03be50d392c166a61e21222c05 (MD5) Previous issue date: 2018-04-17 / Dengue é uma doença infecciosa que afeta países subtropicais. Autoridades de saúde locais utilizam informações sobre o número de notificações para monitorar e prever epidemias. Este trabalho foca na modelagem do número de casos de dengue semanal em quatro cidades do estado do Rio de Janeiro: Rio de Janeiro, São Gonçalo, Campos dos Goytacazes, e Petrópolis. Modelos de séries temporais são frequentemente utilizados para prever o número de casos de dengue nos próximos ciclos (semanas ou meses), particularmente, modelos SARIMA (Modelo Sazonal Autorregressivo Integrado de Médias Móveis) apresentam uma boa performance em situações distintas. Modelagens alternativas ainda incluem informação sobre o clima da região para melhorar a performance preditiva. Apesar disso, modelos que usam apenas dados históricos e de clima podem não possuir informações suficientes para capturar mudanças entre os regimes de não-epidemia e epidemia. Duas razões para isso são o atraso na notificação dos casos e que possivelmente não houveram epidemias nos anos anteriores. Baseando-se no sistema de monitoramento InfoDengue, esperasse que incluindo dados sobre ”numero de reprodução efetiva dos mosquitos”(RT) e ”número de tweets se referindo a dengue”(tweets) possam melhorar a qualidade das previsões no curto (1 semana) e longo (8 semanas) prazo. Foi possível mostrar que modelos de séries temporais incluindo RT e informações climáticas frequentemente performam melhor do que o modelo SARIMA em termos do erro preditivo quadrático médio (RMSE). Incluir a variável sobre o twitter não mostrou uma melhora no RMSE. / Dengue fever is an infectious disease affecting subtropical countries. Local health departments use the number of notified cases to monitor and predict epidemics. This work focus on modeling weekly incidence of dengue fever in four cities of the state of Rio de Janeiro: Rio de Janeiro, São Gonçalo, Campos dos Goytacazes, and Petrópolis. Time series models are often used to predict the number of cases in the next cycles (weeks, months), in particular, SARIMA (Seazonal Auto-Regressive Integrated Moving Average) models are shown to perform well in distinct settings. Alternative models also include climate covariates to improve the quality of the forecasts. However, models that only use historical and climate data may no have sufficient information to capture changes from non-epidemic to an epidemic regime. Two reasons are that there is a delay in the notification of cases and there might not have had epidemics in the previous years. Based on the INFODENGUE monitoring system we argue data including the "effective reproduction number of mosquitoes" (RT) and "number tweets referring to dengue" (tweets) may improve the quality of forecasts in the short (1 week) to long (8 weeks) range. We show that time series models including RT and climate information often outperform SARIMA models in terms of mean squared predictive error (RMSE). Inclusion of twitter did not improve the RMSE.

Today, digitalization is a key factor for businesses to enhance growth and gain advantages and insight in their operations. Both in planning operations and understanding customers the digitalization processes today have key roles, and companies are spending more and more resources in this fields to gain critical insights and enhance growth. The fast-food industry is no exception where restaurants need to be highly flexible and agile in their work. With this, there exists an immense demand for knowledge and insights to help restaurants plan their daily operations and there is a great need for organizations to continuously adapt new technological solutions into their existing processes. Well implemented Machine Learning solutions in combination with feature engineering are likely to bring value into the existing processes. Sales forecasting, which is the main field of study in this thesis work, has a vital role in planning of fast food restaurant's operations, both for budgeting purposes, but also for staffing purposes. The word fast food describes itself. With this comes a commitment to provide high quality food and rapid service to the customers. Understaffing can risk violating either quality of the food or service while overstaffing leads to low overall productivity. Generating highly reliable sales forecasts are thus vital to maximize profits and minimize operational risk. SARIMA, XGBoost and Random Forest were evaluated on training data consisting of sales numbers, business hours and categorical variables describing date and month. These models worked as base learners where sales predictions from a specific dataset were used as training data for a Support Vector Regression model (SVR). A stacking approach to this type of project shows sufficient results with a significant gain in prediction accuracy for all investigated restaurants on a 6-week aggregated timeline compared to the existing solution. / Digitalisering har idag en nyckelroll för att skapa tillväxt och insikter för företag, dessa insikter ger fördelar både inom planering och i förståelsen om deras kunder. Det här är ett område som företag lägger mer och mer resurser på för att skapa större förståelse om sin verksamhet och på så sätt öka tillväxten. Snabbmatsindustrin är inget undantag då restauranger behöver en hög grad av flexibilitet i sina arbetssätt för att möta kundbehovet. Det här skapar en stor efterfrågan av kunskap och insikter för att hjälpa dem i planeringen av deras dagliga arbete och det finns ett stort behov från företagen att kontinuerligt implementera nya tekniska lösningar i befintliga processer. Med väl implementerade maskininlärningslösningar i kombination med att skapa mer informativa variabler från befintlig data kan aktörer skapa mervärde till redan existerande processer. Försäljningsprognostisering, som är huvudområdet för den här studien, har en viktig roll för verksamhetsplaneringen inom snabbmatsindustrin, både inom budgetering och bemanning. Namnet snabbmat beskriver sig själv, med det följer ett löfte gentemot kunden att tillhandahålla hög kvalitet på maten samt att kunna tillhandahålla snabb service. Underbemanning kan riskera att bryta någon av dessa löften, antingen i undermålig kvalitet på maten eller att inte kunna leverera snabb service. Överbemanning riskerar i stället att leda till ineffektivitet i användandet av resurser. Att generera högst tillförlitliga prognoser är därför avgörande för att kunna maximera vinsten och minimera operativ risk. SARIMA, XGBoost och Random Forest utvärderades på ett träningsset bestående av försäljningssiffror, timme på dygnet och kategoriska variabler som beskriver dag och månad. Dessa modeller fungerar som basmodeller vars prediktioner från ett specifikt testset används som träningsdata till en Stödvektorsreggresionsmodell (SVR). Att använda stapling av maskininlärningsmodeller till den här typen av problem visade tillfredställande resultat där det påvisades en signifikant förbättring i prediktionssäkerhet under en 6 veckors aggregerad period gentemot den redan existerande modellen.

[pt] No Brasil, grandes consumidores podem estabelecer seus contratos de energia elétrica em dois ambientes: Ambiente de Contratação Regulado e Ambiente de Contratação Livre. Grandes consumidores são aqueles que possuem carga igual ou superior a 2 MW e podem ser atendidos sob contratos firmados em quaisquer um desses ambientes. Já os consumidores com demanda contratada inferior a 2 MW e superior a 500 kW podem ter seu contrato de energia estabelecido no Ambiente de Contratação Livre proveniente de geração de energia renovável ou no Ambiente de Contratação Regulada através das distribuidoras de energia. A principal vantagem do Ambiente de Contratação Livre é a possibilidade de negociar contratos com diferentes parâmetros, como, por exemplo, preço, quantidade de energia e prazo. Eventuais diferenças entre a energia contratada e a consumida, são liquidadas ao preço de energia de curto prazo, que pode ser bastante volátil.Neste caso o desafio é estabelecer uma estratégia de contratação que minimize os riscos associados a este ambiente. Esta dissertação propõe uma metodologia que envolve a simulação estatística de cenários correlacionados de energia, demanda máxima e preço de curto prazo (também chamado de PLD – Preço de Liquidação das Diferenças) para serem inseridos em um modelo matemático de otimização estocástica, que define os parâmetros ótimos da contratação de energia e demanda. Na parte estatística, um modelo Box e Jenkins é usado para estimar os parâmetros das séries históricas de energia e demanda máxima com o objetivo de simular cenários correlacionados com o PLD. Na parte de otimização, emprega-se uma combinação convexa entre Valor Esperado (VE) e Conditional Value-at-Risk (CVaR) como medidas de risco para encontrar os valores ótimos dos parâmetros contratuais, como a demanda máxima contratada, o volume mensal de energia a ser contratado, além das flexibilidades inferior e superior da energia contratada. Para ilustrar a abordagem proposta, essa metodologia é aplicada a um estudo de caso real para um grande consumidor no Ambiente de Contratação Livre. Os resultados indicaram que a metodologia proposta pode ser uma ferramenta eficiente para consumidores no Ambiente de Contratação Livre e, dado à natureza do modelo, pode ser generalizado para diferentes contratos e mercados de energia. / [en] In Brazil, big consumers can choose their energy contract between two different energy environments: Regulated Contract Environment and Free Contract Environment. Big consumers are characterized by installed load capacity equal or greater than 2 MW and can firm an energy contract under any of these environments. For those consumers with installed load lower than 2 MW and higher than 500 kW, their energy contracts can be firmed in the Free Contract Environment using renewable energy generation or in the Regulated Contract Environment by local distribution companies. The main advantage of the Free Market Environment is the possibility of negotiating contracts with different parameters such as, for example, price, energy quantity and deadlines. Possible differences between contracted energy and consumed energy are settled by the spot price, which can be rather volatile. In this case, the challenge is to establish a contracting strategy that minimize the associated risks with this environment. This thesis proposes a methodology that involves statistical simulation of correlated energy, peak demand and Spot Price scenarios to be used in a stochastic optimization model that defines the optimal energy and demand contract parameters. In the statistical part, a Box and Jenkins model is used to estimate parameters for energy and peak demand in order to simulate scenarios correlated with Spot Price. In the optimization part, a convex combination of Expected Value (EV) and Conditional Value-at-Risk (CVaR) is used as risk measures to find the optimal contract parameters, such as the contracted peak demand, the seasonal energy contracted volumes, in addition to the upper and lower energy contracted bound. To illustrate this approach, this methodology is applied in a real case study for a big consumer with an active Free Market Environment contract. The results indicate that the proposed methodology can be a efficient tool for consumers in the Free Contract Environment and, due to the nature of the model, it can be generalized for different energy contracts and markets.

The interest in modeling non-maturing deposits has skyrocketed ever since thefinancial crisis 2008. Not only from a regulatory and legislative perspective,but also from an investment and funding perspective.Modeling of non-maturing deposits is a very broad subject. In this thesis someof the topics within the subject are investigated, where the greatest focus inon the modeling of the deposit volumes. The main objective is to providethe bank with an analysis of the majority of the topics that needs to be cov-ered when modeling non-maturing deposits. This includes short-rate model-ing using Vasicek’s model, deposit rate modeling using a regression approachand a method proposed by Jarrow and Van Deventer, volume modeling usingSARIMA, SARIMAX and a general additive model, a static replicating port-folio based on Maes and Timmerman’s to model the behaviour of the depositaccounts and finally a liquidity risk model that was suggested by Kalkbrenerand Willing. All of these models have been applied on three different accounttypes: private transaction accounts, savings accounts and corporate savingsaccounts.The results are that, due to the current market, the static replicating portfoliodoes not achieve the desired results. Furthermore, the best volume model forthe data provided is a SARIMA model, meaning the effect of the exogenousvariables are seemingly already embedded in the lagged volume. Finally, theliquidity risk results are plausible and thus deemed satisfactory. / Intresset för att modellera inlåningsvolymer utan en kontrakterad förfallodaghar ökat markant sedan finanskrisen 2008. Inte bara sett utifrån ett perspek-tiv att uppfylla krav som ställs av tillsynsmyndigheter, men också sett utifrånbankens investerings-och finansieringsperspektiv.Målet med det här arbetet är att förse banken med en analys av majoritetenav de olika områdena som man behöver ta hänsyn till när man ska model-lera inlåningar utan förfallodatum, men med ett fokus på volymmodellering.I den här rapporten modelleras räntor (kortränta och kontoränta), kontovoly-merna, kontobeteendet samt likviditetsrisken. Detta görs med hjälp av Vasicekför korträntan, en regressionsmetod samt en metod som föreslagits av Jarrowoch Van Deventer för kontoräntan, SARIMA, SARIMAX och en generell ad-ditiv regressionsmetod för volymerna, en statisk replikeringsportfölj baseradpå Maes och Timmermans modell för att imitera kontona och slutligen så mo-delleras likviditetsrisken med ett ramverk som föreslagits av Kalkbrener ochWilling. Alla dessa nämnda modeller appliceras, där det är möjligt, på de treolika kontotyperna: privatkonton, sparkonton samt företagssparkonto.Resultatet är att räntemodelleringen samt replikeringsportföljen inte ger ade-kvata resultat på grund av den rådande marknaden. Vidare så ger en SARIMA-modell den bästa prediktionen, vilket gör att slutsatsen är att andra exogenavariabler redan är inneslutna i den fördröjda volymvariabeln. Avslutningsvisså ger likviditetsmodellen tillfredsställande resultat och antas vara rimlig.

Submitted by Yuri Barboza Cocchiarale (yuriufrj1410@hotmail.com) on 2017-07-18T13:48:06Z No. of bitstreams: 1 Dissertação FGV - Yuri Barboza 2017.pdf: 7367125 bytes, checksum: f7c2e1f50effbdca635ca98fe61d1862 (MD5) / Approved for entry into archive by GILSON ROCHA MIRANDA (gilson.miranda@fgv.br) on 2017-07-21T14:50:15Z (GMT) No. of bitstreams: 1 Dissertação FGV - Yuri Barboza 2017.pdf: 7367125 bytes, checksum: f7c2e1f50effbdca635ca98fe61d1862 (MD5) / Made available in DSpace on 2017-07-27T12:28:16Z (GMT). No. of bitstreams: 1 Dissertação FGV - Yuri Barboza 2017.pdf: 7367125 bytes, checksum: f7c2e1f50effbdca635ca98fe61d1862 (MD5) Previous issue date: 2017-05-31 / The aim of this study is to show that the creation of the Law no. 12,858 / 2013 has an irrelevant effect in the short term, regarding issues related to health and education problems. By adopting some assumptions capable of modeling the obtained database and estimating oil production using the SARIMA, Holt-Winters and Kalman Filter models, combined with the price forecast, the portrayed results can consistently reflect the impacts brought by this law. The problem becomes even bigger in a long-term horizon, in which the migration of revenue from royalties will overwhelmingly affect the entities who benefit from it. / Este trabalho tem como objetivo mostrar que a criação da Lei n° 12.858/2013 possui um efeito irrelevante no curto prazo, no que tange as questões ligadas aos problemas da saúde e educação. Adotando algumas premissas capazes de modelar o banco de dados obtido e estimando a produção do petróleo utilizando os modelos SARIMA, Holt-Winters e Filtro de Kalman, combinadas com a previsão dos preços, os resultados apresentados conseguem refletir de forma consistente os impactos trazidos por esta lei. O problema ainda se torna maior em um horizonte de longo prazo, onde a migração da receita provida dos royalties afetará de forma devastadora os entes que se beneficiam dela.

Submitted by Basiliki Theophane Calochorios Litvac (basiliki.litvac@gmail.com) on 2013-03-06T22:27:11Z No. of bitstreams: 1 dissertacao_Basiliki_final_rev.pdf: 681459 bytes, checksum: 86dfce2ca595dfd933509d266c084a2d (MD5) / Approved for entry into archive by Suzinei Teles Garcia Garcia (suzinei.garcia@fgv.br) on 2013-03-07T12:53:10Z (GMT) No. of bitstreams: 1 dissertacao_Basiliki_final_rev.pdf: 681459 bytes, checksum: 86dfce2ca595dfd933509d266c084a2d (MD5) / Made available in DSpace on 2013-03-07T13:14:10Z (GMT). No. of bitstreams: 1 dissertacao_Basiliki_final_rev.pdf: 681459 bytes, checksum: 86dfce2ca595dfd933509d266c084a2d (MD5) Previous issue date: 2013-02-05 / Este trabalho tem por objetivo avaliar para o caso brasileiro uma das mais importantes propriedades esperadas de um núcleo: ser um bom previsor da inflação plena futura. Para tanto, foram utilizados como referência para comparação dois modelos construídos a partir das informações mensais do IPCA e seis modelos VAR referentes a cada uma das medidas de núcleo calculadas pelo Banco Central do Brasil. O desempenho das previsões foi avaliado pela comparação dos resultados do erro quadrático médio e pela aplicação da metodologia de Diebold-Mariano (1995) de comparação de modelos. Os resultados encontrados indicam que o atual conjunto de medidas de núcleos calculado pelo Banco Central não atende pelos critérios utilizados neste trabalho a essa característica desejada. / This paper aims at evaluating one of the most important desirable properties of a core inflation measure: to be a better predictor of headline inflation over the future. To achieve this goal, two benchmark models using monthly IPCA data were compared with six VAR models for each one of the core measures calculated by the Brazilian Central Bank. The forecasting performance was evaluated comparing the mean square error and by the Diebold-Mariano (1995) test for predictive accuracy. The evidence found indicates that the current set of core inflation measures calculated by the Central Bank does fulfill this desired property.

Submitted by Simone Jager (si_jager@hotmail.com) on 2014-02-10T15:30:57Z No. of bitstreams: 1 Simone Jager 2014.pdf: 764649 bytes, checksum: 100e29a7572ff1d6c57a770ace28e1bf (MD5) / Approved for entry into archive by Vitor Souza (vitor.souza@fgv.br) on 2014-02-24T21:08:40Z (GMT) No. of bitstreams: 1 Simone Jager 2014.pdf: 764649 bytes, checksum: 100e29a7572ff1d6c57a770ace28e1bf (MD5) / Made available in DSpace on 2014-05-20T13:15:26Z (GMT). No. of bitstreams: 1 Simone Jager 2014.pdf: 764649 bytes, checksum: 100e29a7572ff1d6c57a770ace28e1bf (MD5) Previous issue date: 2014-01-23 / This paper compares time series models to forecast short-term Brazilian inflation measured by Consumer Price Index (IPCA). Were considered SARIMA Box-Jenkins models and structural models in state space, as estimated by the Kalman filter. For estimation of the models, the series of IPCA monthly basis from March 2003 to March 2012 was used. The SARIMA models were estimated in EVIEWS and structural models in STAMP. For the validation of the models out of sample forecasts were considered one step ahead for the period April 2012 to March 2013, based on the main criteria for assessing predictive ability proposed in the literature. The conclusion of the study is that, although the structural model allows, to decompose the series into components with direct interpretation and study them separately, while incorporating explanatory variables in a simple way, the performance of the SARIMA model to predict Brazilian inflation was higher in the period and horizon considered. Another important positive aspect is that the implementation of a SARIMA model is ready, and predictions from it are obtained in a simple and direct way. / Este trabalho compara modelos de séries temporais para a projeção de curto prazo da inflação brasileira, medida pelo Índice de Preços ao Consumidor Amplo (IPCA). Foram considerados modelos SARIMA de Box e Jenkins e modelos estruturais em espaço de estados, estimados pelo filtro de Kalman. Para a estimação dos modelos, foi utilizada a série do IPCA na base mensal, de março de 2003 a março de 2012. Os modelos SARIMA foram estimados no EVIEWS e os modelos estruturais no STAMP. Para a validação dos modelos para fora da amostra, foram consideradas as previsões 1 passo à frente para o período de abril de 2012 a março de 2013, tomando como base os principais critérios de avaliação de capacidade preditiva propostos na literatura. A conclusão do trabalho é que, embora o modelo estrutural permita, decompor a série em componentes com interpretação direta e estudá-las separadamente, além de incorporar variáveis explicativas de forma simples, o desempenho do modelo SARIMA para prever a inflação brasileira foi superior, no período e horizonte considerados. Outro importante aspecto positivo é que a implementação de um modelo SARIMA é imediata, e previsões a partir dele são obtidas de forma simples e direta.

This work aims to explore practical one-step-ahead forecasting of structurally changing data, an unstable behaviour, that real-life data connected to human activity often exhibit. This setting can be characterized as monitoring process. Various forecast models, methods and approaches can range from being simple and computationally "cheap" to very sophisticated and computationally "expensive". Moreover, different forecast methods handle different data-patterns and structural changes differently: for some particular data types or data intervals some particular forecast methods are better than the others, something that is usually not known beforehand. This raises a question: "Can one design a forecast procedure, that effectively and optimally switches between various forecast methods, adapting the forecast methods usage to the changes in the incoming data flow?" The thesis answers this question by introducing optimality concept, that allows optimal switching between simultaneously executed forecast methods, thus "tailoring" forecast methods to the changes in the data. It is also shown, how another forecast approach: combinational forecasting, where forecast methods are combined using weighted average, can be utilized by optimality principle and can therefore benefit from it. Thus, four classes of forecast results can be considered and compared: basic forecast methods, basic optimality, combinational forecasting, and combinational optimality. The thesis shows, that the usage of optimality gives results, where most of the time optimality is no worse or better than the best of forecast methods, that optimality is based on. Optimality reduces also scattering from multitude of various forecast suggestions to a single number or only a few numbers (in a controllable fashion). Optimality gives additionally lower bound for optimal forecasting: the hypothetically best achievable forecast result. The main conclusion is that optimality approach makes more or less obsolete other traditional ways of treating the monitoring processes: trying to find the single best forecast method for some structurally changing data. This search still can be sought, of course, but it is best done within optimality approach as its innate component. All this makes the proposed optimality approach for forecasting purposes a valid "representative" of a more broad ensemble approach (which likewise motivated development of now popular Ensemble Learning concept as a valid part of Machine Learning framework). / Denna avhandling syftar till undersöka en praktisk ett-steg-i-taget prediktering av strukturmässigt skiftande data, ett icke-stabilt beteende som verkliga data kopplade till människoaktiviteter ofta demonstrerar. Denna uppsättning kan alltså karakteriseras som övervakningsprocess eller monitoringsprocess. Olika prediktionsmodeller, metoder och tillvägagångssätt kan variera från att vara enkla och "beräkningsbilliga" till sofistikerade och "beräkningsdyra". Olika prediktionsmetoder hanterar dessutom olika mönster eller strukturförändringar i data på olika sätt: för vissa typer av data eller vissa dataintervall är vissa prediktionsmetoder bättre än andra, vilket inte brukar vara känt i förväg. Detta väcker en fråga: "Kan man skapa en predictionsprocedur, som effektivt och på ett optimalt sätt skulle byta mellan olika prediktionsmetoder och för att adaptera dess användning till ändringar i inkommande dataflöde?" Avhandlingen svarar på frågan genom att introducera optimalitetskoncept eller optimalitet, något som tillåter ett optimalbyte mellan parallellt utförda prediktionsmetoder, för att på så sätt skräddarsy prediktionsmetoder till förändringar i data. Det visas också, hur ett annat prediktionstillvägagångssätt: kombinationsprediktering, där olika prediktionsmetoder kombineras med hjälp av viktat medelvärde, kan utnyttjas av optimalitetsprincipen och därmed få nytta av den. Alltså, fyra klasser av prediktionsresultat kan betraktas och jämföras: basprediktionsmetoder, basoptimalitet, kombinationsprediktering och kombinationsoptimalitet. Denna avhandling visar, att användning av optimalitet ger resultat, där optimaliteten för det mesta inte är sämre eller bättre än den bästa av enskilda prediktionsmetoder, som själva optimaliteten är baserad på. Optimalitet reducerar också spridningen från mängden av olika prediktionsförslag till ett tal eller bara några enstaka tal (på ett kontrollerat sätt). Optimalitet producerar ytterligare en nedre gräns för optimalprediktion: det hypotetiskt bästa uppnåeliga prediktionsresultatet. Huvudslutsatsen är följande: optimalitetstillvägagångssätt gör att andra traditionella sätt att ta hand om övervakningsprocesser blir mer eller mindre föråldrade: att leta bara efter den enda bästa enskilda prediktionsmetoden för data med strukturskift. Sådan sökning kan fortfarande göras, men det är bäst att göra den inom optimalitetstillvägagångssättet, där den ingår som en naturlig komponent. Allt detta gör det föreslagna optimalitetstillvägagångssättetet för prediktionsändamål till en giltig "representant" för det mer allmäna ensembletillvägagångssättet (något som också motiverade utvecklingen av numera populär Ensembleinlärning som en giltig del av Maskininlärning).