Página 1 dos resultados de 210 itens digitais encontrados em 0.008 segundos

The combination of neural estimates in prediction and decision problems

Freitas, Paulo Sérgio Abreu
Fonte: Universidade de Lisboa: Faculdade de Ciências Publicador: Universidade de Lisboa: Faculdade de Ciências
Tipo: Tese de Doutorado
Publicado em //2008 ENG
Relevância na Pesquisa
35.94%
In this dissertation, different ways of combining neural predictive models or neural-based forecasts are discussed. The proposed approaches consider mostly Gaussian radial basis function networks, which can be efficiently identified and estimated through recursive/adaptive methods. Two different ways of combining are explored to get a final estimate – model mixing and model synthesis –, with the aim of obtaining improvements both in terms of efficiency and effectiveness. In the context of model mixing, the usual framework for linearly combining estimates from different models is extended, to deal with the case where the forecast errors from those models are correlated. In the context of model synthesis, and to address the problems raised by heavily nonstationary time series, we propose hybrid dynamic models for more advanced time series forecasting, composed of a dynamic trend regressive model (or, even, a dynamic harmonic regressive model), and a Gaussian radial basis function network. Additionally, using the model mixing procedure, two approaches for decision-making from forecasting models are discussed and compared: either inferring decisions from combined predictive estimates, or combining prescriptive solutions derived from different forecasting models. Finally...

Previsão de arrecadação tributária baseada em um método de otimização de portfólio para a combinação de previsões; Revenue forecast based on a portfolio optimization method for combination of forecasts

Kubo, Sergio Hideo
Fonte: Biblioteca Digitais de Teses e Dissertações da USP Publicador: Biblioteca Digitais de Teses e Dissertações da USP
Tipo: Tese de Doutorado Formato: application/pdf
Publicado em 01/08/2014 PT
Relevância na Pesquisa
36.46%
Uma previsão de receitas precisa é muito importante para o administrador público na elaboração do orçamento anual, e para isso há a necessidade de se encontrar um modelo, econométrico ou não, que possibilite essa previsão com qualidade. Este trabalho apresenta uma forma inovadora para realizar a combinação de modelos de previsão. Seu objetivo foi criar uma metodologia para a obtenção de pesos para a combinação de modelos baseada no método de otimização de uma carteira de investimentos proposto por Markowitz. Para o estudo, foram utilizadas as estimações de três a cinco previsões individuais de um a cinco passos à frente, com os modelos Box-Jenkins SARIMA (Autorregressivo Integrado de Médias Móveis Sazonal), PLSR (Regressão com Mínimos Quadrados Parciais) e o Método não econométrico de Indicadores, como é denominado internamente na Receita Federal. A utilização da fronteira eficiente de Markowitz, que apresenta os pontos de mínima variância para cada retorno, é semelhante à minimização da variância da combinação, proposta no artigo seminal de Bates e Granger. O risco (desvio padrão), na teoria de portfólio de Markowitz, pode ser definido como a dispersão dos resultados e pode ser decomposto em risco sistemático e risco não sistemático. À medida que a quantidade de pesos das previsões a combinar cresce...

Using Common Features to Understand the Behavior of Metal-Commodity Prices and Forecast them at Different Horizons

Issler, João Victor; Rodrigues, Claudia; Burjack, Rafael
Fonte: Escola de Pós-Graduação em Economia da FGV Publicador: Escola de Pós-Graduação em Economia da FGV
Tipo: Relatório
EN_US
Relevância na Pesquisa
36.24%
The objective of this article is to study (understand and forecast) spot metal price levels and changes at monthly, quarterly, and annual frequencies. Data consists of metal-commodity prices at a monthly and quarterly frequencies from 1957 to 2012, extracted from the IFS, and annual data, provided from 1900-2010 by the U.S. Geological Survey (USGS). We also employ the (relatively large) list of co-variates used in Welch and Goyal (2008) and in Hong and Yogo (2009). We investigate short- and long-run comovement by applying the techniques and the tests proposed in the common-feature literature. One of the main contributions of this paper is to understand the short-run dynamics of metal prices. We show theoretically that there must be a positive correlation between metal-price variation and industrial-production variation if metal supply is held fixed in the short run when demand is optimally chosen taking into account optimal production for the industrial sector. This is simply a consequence of the derived-demand model for cost-minimizing firms. Our empirical evidence fully supports this theoretical result, with overwhelming evidence that cycles in metal prices are synchronized with those in industrial production. This evidence is stronger regarding the global economy but holds as well for the U.S. economy to a lesser degree. Regarding out-of-sample forecasts...

Improvement on the sales forecast accuracy for a fast growing company by the best combination of historical data usage and clients segmentation

Burgada Muñoz, Santiago
Fonte: Fundação Getúlio Vargas Publicador: Fundação Getúlio Vargas
Tipo: Dissertação
EN_US
Relevância na Pesquisa
36.43%
Industrial companies in developing countries are facing rapid growths, and this requires having in place the best organizational processes to cope with the market demand. Sales forecasting, as a tool aligned with the general strategy of the company, needs to be as much accurate as possible, in order to achieve the sales targets by making available the right information for purchasing, planning and control of production areas, and finally attending in time and form the demand generated. The present dissertation uses a single case study from the subsidiary of an international explosives company based in Brazil, Maxam, experiencing high growth in sales, and therefore facing the challenge to adequate its structure and processes properly for the rapid growth expected. Diverse sales forecast techniques have been analyzed to compare the actual monthly sales forecast, based on the sales force representatives’ market knowledge, with forecasts based on the analysis of historical sales data. The dissertation findings show how the combination of both qualitative and quantitative forecasts, by the creation of a combined forecast that considers both client´s demand knowledge from the sales workforce with time series analysis, leads to the improvement on the accuracy of the company´s sales forecast.

Modelo composto para prever demanda através da integração de previsões; Composed model to foresee demand through the integration of forecasts

Werner, Liane; Ribeiro, Jose Luis Duarte
Fonte: Universidade Federal do Rio Grande do Sul Publicador: Universidade Federal do Rio Grande do Sul
Tipo: Artigo de Revista Científica Formato: application/pdf
POR
Relevância na Pesquisa
36.35%
Realizar previsões de demanda é uma atividade importante na empresa, entretanto, usar uma única técnica para obtê-las pode não ser suficiente para incorporar todo o conhecimento associado ao ambiente de previsão. As formas de integração de previsões incorporam várias técnicas e têm mostrado potencial para reduzir o erro de previsão. Este trabalho apresenta uma modelagem que está estruturada utilizando: combinação de previsões e ajuste baseado na opinião. Os elementos incluídos na modelagem são: dados históricos; econômicos; e de especialistas. Após obter-se a previsão combinada, aplica-se um ajuste para obter a previsão final. O modelo proposto é ilustrado através de uma aplicação.; Demand forecasting is an important task in the companies, however the use of a single technique to produce forecasts might not be enough to gather all the knowledge associated with the forecast environment. The way to integrate forecasts incorporates various techniques and has show potential to reduce forecast error. This study presents a model that relies on the use of two means of integration: forecast combination and judgmental adjustment. The elements covered by the presented model are: historic data, economic data, and the opinion of experts. After obtaining the combined forecast...

Properties advanced of the silicon nitride based ceramics and recent performance on automotive parts manufacture by machining process: Advanced Ceramics, demand Forecast

De Souza, José Vitor Candido; De Andrade Nono, Maria Do Carmo; Mineiro, Sergio Luiz; De MacEdo Silva, Olivério Moreira; Ribeiro, Marcos Valério
Fonte: Universidade Estadual Paulista Publicador: Universidade Estadual Paulista
Tipo: Conferência ou Objeto de Conferência
ENG
Relevância na Pesquisa
36.03%
Automotive parts manufacture by machining process using silicon nitride-based ceramic tool development in Brazil already is a reality. Si 3N4-based ceramic cutting tools offer a high productivity due to their excellent hot hardness, which allows high cutting speeds. Under such conditions the cutting tool must be resistant to a combination of mechanical, thermal and chemical attacks. Silicon nitride based ceramic materials constitute a mature technology with a very broad base of current and potential applications. The best opportunities for Si3N 4-based ceramics include ballistic armor, composite automotive brakes, diesel particulate filters, joint replacement products and others. The goal of this work was to show latter advance in silicon nitride manufacture and its recent evolution on machining process of gray cast iron, compacted graphite iron and Ti-6Al-4V. Materials characterization and machining tests were analyzed by X-Ray Diffraction, Scanning Electron Microscopy, Vickers hardness and toughness fracture and technical norm. In recent works the authors has been proved to advance in microstructural, mechanical and physic properties control. These facts prove that silicon nitride-based ceramic has enough resistance to withstand the impacts inherent to the machining of gray cast iron (CI)...

Combinações de modelos de previsão da estrutura a termo da taxa de juros : aplicações ao caso brasileiro

Araújo, Rafael Cavalcanti de
Fonte: Universidade de Brasília Publicador: Universidade de Brasília
Tipo: Dissertação
POR
Relevância na Pesquisa
26.45%
Dissertação (mestrado)—Universidade de Brasília, Departamento de Economia, 2011.; Questões como quebras estruturais e problemas de especificação tornam difícil a possibilidade de existência de um modelo de previsão da estrutura a termo da taxa de juros que domine os demais competidores. Essa dissertação tem como objetivo principal identificar a existência de métodos de combinação que permitam obter previsões da curva de juros superiores às dos modelos individuais para o caso brasileiro.A análise dos resultados dos modelos individuais confirma não ser possível encontrar um modelo que consistentemente produza menor erro de previsão. Adicionalmente, os desempenhos relativos podem variar ao longo do tempo. Os problemas de utilização de modelos individuais podem ser reduzidos pelo intermédio de métodos de combinação de previsões. Os resultados alcançados para o período analisado permitem atestar ganhos de previsão consistentes ao longo tempo. Em particular, quanto mais longo o horizonte considerado, maior a contribuição das combinações. _______________________________________________________________________________ ABSTRACT; Issues like structural breaks and misspecification make it difficult to find any term structure of interest rates model that dominates all competitors. This dissertation aims to identify the existence of combination methods that produce superior forecast results compared to individual models applied to Brazilian data. Empirical results confirm not being possible to determine a individual model capable of consistently produce superior forecasts. Furthermore...

Skill assessment for an operational algal bloom forecast system

Stumpf, Richard P.; Tomlinson, Michelle C.; Calkins, Julie A.; Kirkpatrick, Barbara; Fisher, Kathleen; Nierenberg, Kate; Currier, Robert; Wynne, Timothy T.
Fonte: PubMed Publicador: PubMed
Tipo: Artigo de Revista Científica
Publicado em 20/02/2009 EN
Relevância na Pesquisa
26.27%
An operational forecast system for harmful algal blooms (HABs) in southwest Florida is analyzed for forecasting skill. The HABs, caused by the toxic dinoflagellate, Karenia brevis, lead to shellfish toxicity and to respiratory irritation. In addition to predicting new blooms and their extent, HAB forecasts are made twice weekly during a bloom event, using a combination of satellite derived image products, wind predictions, and a rule-based model derived from previous observations and research. These forecasts include: identification, intensification, transport, extent, and impact; the latter being the most significant to the public. Identification involves identifying new blooms as HABs and is validated against an operational monitoring program involving water sampling. Intensification forecasts, which are much less frequently made, can only be evaluated with satellite data on mono-specific blooms. Extent and transport forecasts of HABs are also evaluated against the water samples. Due to the resolution of the forecasts and available validation data, skill cannot be resolved at scales finer than 30 km. Initially, respiratory irritation forecasts were analyzed using anecdotal information, the only available data, which had a bias toward major respiratory events leading to a forecast accuracy exceeding 90%. When a systematic program of twice-daily observations from lifeguards was implemented...

Evaluation of Northwest Pacific tropical cyclone track forecast difficulty and skill as a function of environmental structure

Webb, Benny H.
Fonte: Monterey, California. Naval Postgraduate School Publicador: Monterey, California. Naval Postgraduate School
Tipo: Tese de Doutorado
EN_US
Relevância na Pesquisa
26.24%
A Systematic Approach for tropical cyclone track forecasting by Carr and Elsberry defines the Synoptic Environment of each cyclone in terms of ten Synoptic Pattern/Region combinations. Because storms in each Pattern/Region combination have characteristic tracks that are dramatically different, it is hypothesized that the degree of difficulty in forecasting the tropical cyclone track, and the skill of the Joint Typhoon Warning Center (JTWC) track forecasts will be a function of the Synoptic Environment. The degree of forecast difficulty is defined by comparing forecast track errors (FTEs) of the operational CLImatology and PERsistence (CLIPER) technique in each of the ten Pattern/Region combinations with the overall CLIPER FTEs. The most difficult combinations are the recurving scenarios of Weakened Ridge Region of the Standard Pattern and the Southerly Flow Region of the Multiple tropical cyclone Pattern. The least difficult combinations are the Dominant Ridge Regions of the Standard and Gyre Patterns. The JTWC forecasts have statistically significant skill compared to the no-skill CLIPER forecasts for storms in the Standard/Dominant Ridge and North-oriented Pattern/North-Oriented Region, which comprise nearly 77% of the five-year sample of JTWC forecasts. As transitions occur between the Synoptic Pattern/Region combinations...

Projeção de preços de alumínio: modelo ótimo por meio de combinação de previsões; Aluminum price forecasting: optimal forecast combination

Castro, João Bosco Barroso de
Fonte: Biblioteca Digitais de Teses e Dissertações da USP Publicador: Biblioteca Digitais de Teses e Dissertações da USP
Tipo: Tese de Doutorado Formato: application/pdf
Publicado em 15/06/2015 PT
Relevância na Pesquisa
66.55%
Commodities primárias, tais como metais, petróleo e agricultura, constituem matérias-primas fundamentais para a economia mundial. Dentre os metais, destaca-se o alumínio, usado em uma ampla gama de indústrias, e que detém o maior volume de contratos na London Metal Exchange (LME). Como o preço não está diretamente relacionado aos custos de produção, em momentos de volatilidade ou choques econômicos, o impacto financeiro na indústria global de alumínio é significativo. Previsão de preços do alumínio é fundamental, portanto, para definição de política industrial, bem como para produtores e consumidores. Este trabalho propõe um modelo ótimo de previsões para preços de alumínio, por meio de combinações de previsões e de seleção de modelos através do Model Confidence Set (MCS), capaz de aumentar o poder preditivo em relação a métodos tradicionais. A abordagem adotada preenche uma lacuna na literatura para previsão de preços de alumínio. Foram ajustados 5 modelos individuais: AR(1), como benchmarking, ARIMA, dois modelos ARIMAX e um modelo estrutural, utilizando a base de dados mensais de janeiro de 1999 a setembro de 2014. Para cada modelo individual, foram geradas 142 previsões fora da amostra, 12 meses à frente...

The pairwise approach to model a large set of disaggregates with common trends

Carlomagno, Guillermo; Espasa, Antoni
Fonte: Universidade Carlos III de Madrid Publicador: Universidade Carlos III de Madrid
Tipo: info:eu-repo/semantics/draft; info:eu-repo/semantics/workingPaper
Publicado em /05/2014 ENG
Relevância na Pesquisa
36.19%
The objective of this paper is to model and forecast all the components of a macro orbusiness variable. Our contribution concerns cases with a large number (hundreds) ofcomponents where multivariate approaches are not feasible. We extend in several directions the pairwise approach originally proposed by Espasa and Mayo-Burgos(2013) and study its statistical properties. The pairwise approach consists on performing common features tests between the N(N-1)/2 pairs of series that exist in a group of N of them. Once this is done, groups of series that share common features can be formed. Next, all the components are forecast using single equation models that include the restrictions derived by the common features. In this paper we focus on discovering groups of components that share single common trends. The asymptotic properties of the procedure are studied analytically. Monte Carlo evidence on the small samples performance is provided and a small samples correction procedure designed. A comparison with a DFM alternative is also carried out, and results indicate that the pairwise approach dominates in many empirically relevant situations. A relevant advantage of the pairwise approach is that it does not need common features to be pervasive. A strategy for dealing with outliers and breaks in the context of the pairwise procedure is designed and its properties studied by Monte Carlo. Results indicate that the treatment of these observations may considerably improve the procedure's performance when series are 'contaminated'.

Estimation and study of forecast error covariances using an ensemble method in a global NWP model; Utilisation d'une méthode d'ensemble pour estimer et étudier les covariances d'erreur dans un modèle météorologique global; Utilização de um método de ensemble para estimar e estudar as estatísticas dos erros de um modelo de previsão numérica do tempo

Pereira, Margarida Belo
Fonte: Universidade de Évora Publicador: Universidade de Évora
Tipo: Tese de Doutorado
ENG
Relevância na Pesquisa
36.19%
Synthesis - The production of an accurate analysis is one important goal of modern NWP centers, where sophisticated data assimilation techniques (such as variational methods) have been implemented. However, this task is not straightforward, since the real state of the atmosphere is never exactly known. One of the main difficulties in data assimilation is caused by the fact that the degrees of freedom of the modern NWP models (~107) are larger than the number of independent available observations (~105). Moreover, the distribution of the observation network is not uniform in space and in time. For these reasons, it is not enough to perform a spatial interpolation of observations into a regular grid. A prior information is needed in order to solve the undetermined analysis problem. In other words, it is necessary to have a first guess about the atmospheric state at all grid points. In modern data assimilation schemes, this first guess (or background) is provided by a short range forecast (from a previous analysis cycle). Hence, the analysis field results from a combination of observations and a background field. The relative weights given to observations and to the background depend on specified ob-servation and background error covariance matrices (which are usually noted R and B...

Combining forecast densities from VARs with uncertain instabilities

Sofie Jore, Anne; Mitchell, James; Vahey, Shaun
Fonte: John Wiley & Sons Inc Publicador: John Wiley & Sons Inc
Tipo: Artigo de Revista Científica
Relevância na Pesquisa
36.19%
Recursive-weight forecast combination is often found to an ineffective method of improving point forecast accuracy in the presence of uncertain instabilities. We examine the effectiveness of this strategy for forecast densities using (many) vector autoreg

Forecast Combination Under Heavy-Tailed Errors

Cheng, Gang; Wang, Sicong; Yang, Yuhong
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Publicado em 26/08/2015
Relevância na Pesquisa
46.55%
Forecast combination has been proven to be a very important technique to obtain accurate predictions. In many applications, forecast errors exhibit heavy tail behaviors for various reasons. Unfortunately, to our knowledge, little has been done to deal with forecast combination for such situations. The familiar forecast combination methods such as simple average, least squares regression, or those based on variance-covariance of the forecasts, may perform very poorly. In this paper, we propose two nonparametric forecast combination methods to address the problem. One is specially proposed for the situations that the forecast errors are strongly believed to have heavy tails that can be modeled by a scaled Student's t-distribution; the other is designed for relatively more general situations when there is a lack of strong or consistent evidence on the tail behaviors of the forecast errors due to shortage of data and/or evolving data generating process. Adaptive risk bounds of both methods are developed. Simulations and a real example show superior performance of the new methods.

On the Forecast Combination Puzzle

Qian, Wei; Rolling, Craig A.; Cheng, Gang; Yang, Yuhong
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Publicado em 03/05/2015
Relevância na Pesquisa
46.5%
It is often reported in forecast combination literature that a simple average of candidate forecasts is more robust than sophisticated combining methods. This phenomenon is usually referred to as the "forecast combination puzzle". Motivated by this puzzle, we explore its possible explanations including estimation error, invalid weighting formulas and model screening. We show that existing understanding of the puzzle should be complemented by the distinction of different forecast combination scenarios known as combining for adaptation and combining for improvement. Applying combining methods without consideration of the underlying scenario can itself cause the puzzle. Based on our new understandings, both simulations and real data evaluations are conducted to illustrate the causes of the puzzle. We further propose a multi-level AFTER strategy that can integrate the strengths of different combining methods and adapt intelligently to the underlying scenario. In particular, by treating the simple average as a candidate forecast, the proposed strategy is shown to avoid the heavy cost of estimation error and, to a large extent, solve the forecast combination puzzle.

Forecasting Distributions with Experts Advice

Sancetta, Alessio
Fonte: Universidade de Cambridge Publicador: Universidade de Cambridge
Formato: 432791 bytes; application/pdf; application/pdf
EN_GB
Relevância na Pesquisa
36.24%
This paper considers forecasts of the distribution of data whose distribution function is possibly time varying. The forecast is achieved via time varying combinations of experts? forecasts. We derive theoretical worse case bounds for general algorithms based on multiplicative updates of the combination weights. The bounds are useful to study the properties of forecast combinations when data are nonstationary and there is no unique best model. An application with an empirical study is used to highlight the results in practice.

Online Forecast Combination for Dependent Heterogeneous Data

Sancetta, Alessio
Fonte: Faculty of Economics, University of Cambridge, UK Publicador: Faculty of Economics, University of Cambridge, UK
Tipo: Trabalho em Andamento
EN
Relevância na Pesquisa
56.2%
This paper studies a procedure to combine individual forecasts that achieve theoretical optimal performance. The results apply to a wide variety of loss functions and no conditions are imposed on the data sequences and the individual forecasts apart from a tail condition. The theoretical results show that the bounds are also valid in the case of time varying combination weights, under specific conditions on the nature of time variation. Some experimental evidence to confirm the results is provided.

Assessing forecast uncertainties in a VECX model for Switzerland: an exercise in forecast combination across models and observation windows

Assenmacher-Wesche, Katrin; Pesaran, M. Hashem
Fonte: Faculty of Economics, University of Cambridge, UK Publicador: Faculty of Economics, University of Cambridge, UK
Tipo: Trabalho em Andamento
EN
Relevância na Pesquisa
36.24%
model for Switzerland. Forecast uncertainty is evaluated in three different dimensions. First, we investigate the effect on forecasting performance of averaging over forecasts from different models. Second, we look at different estimation windows. We find that averaging over estimation windows is at least as effective as averaging over different models and both complement each other. Third, we explore whether using weighting schemes from the machine learning literature improves the average forecast. Compared to equal weights the effect of the weighting scheme on forecast accuracy is small in our application.

Evaluation and combination of conditional quantile forecasts

Giacomini, Raffaella; Komunjer, Ivana
Fonte: Instituto de Tecnologia da Califórnia Publicador: Instituto de Tecnologia da Califórnia
Tipo: Article; PeerReviewed Formato: application/pdf
Publicado em /10/2005
Relevância na Pesquisa
35.94%
We propose an encompassing test for comparing conditional quantile forecasts in an out-of-sample framework. Our test provides a basis for forecast combination when encompassing is rejected. Its central features are (1) use of the "tick" loss function, (2) a conditional approach to out-of-sample evaluation, and (3) derivation in an environment with asymptotically nonvanishing estimation uncertainty. Our approach is valid under general conditions; the forecasts can be based on nested or nonnested models and can be obtained by general estimation procedures. We illustrate the test properties in a Monte Carlo experiment and apply it to evaluate and compare four popular value-at-risk models.

Prevendo o crescimento da produção industrial usando um número limitado de combinações de previsões

Hollauer, Gilberto; Issler, João Victor; Notini, Hilton H.
Fonte: Universidade de São Paulo. Faculdade de Economia, Administração e Contabilidade de RP Publicador: Universidade de São Paulo. Faculdade de Economia, Administração e Contabilidade de RP
Tipo: info:eu-repo/semantics/article; info:eu-repo/semantics/publishedVersion; ; ; Formato: application/pdf
Publicado em 01/01/2008 POR
Relevância na Pesquisa
36.03%
O objetivo central deste artigo é o de propor e avaliar modelos econométricos de previsão para o PIB industrial brasileiro. Para tanto, foram utilizados diversos modelos de previsão como também combinações de modelos. Foi realizada uma análise criteriosa das séries a serem utilizadas na previsão. Nós concluímos que a utilização de vetores de cointegração melhora substancialmente a performance da previsão. Além disso, os modelos de combinação de previsão, na maioria dos casos, tiveram uma performance superior aos demais modelos, que já apresentavam boa capacidade preditiva.; The purpose of this article is to propose and evaluate forecasting models for the Brazilian industrial GDP. Most models are based on vector auto-regressions (VARs) or on restricted VARs, but models on the ARMA class are also entertained. We used many forecasting models and also combinations of these models. The use of cointegration vectors improves substantially the forecast performance of industrial GDP. Furthermore, in general, combining models out-performed individual models, even when the performance of the later was acceptable.