31 |
Máquinas desorganizadas para previsão de séries de vazões / Unorganized machines to seasonal streamflow series forecastingSiqueira, Hugo Valadares, 1983- 24 August 2018 (has links)
Orientadores: Christiano Lyra Filho, Romis Ribeiro de Faissol Attux / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-24T05:06:09Z (GMT). No. of bitstreams: 1
Siqueira_HugoValadares_D.pdf: 10867937 bytes, checksum: 512652380d6dd25b8717bfd5c8f5f0f8 (MD5)
Previous issue date: 2013 / Resumo: Este trabalho explora a possibilidade de aplicação de arquiteturas de redes neurais artificiais - redes neurais de estado de eco (ESN) e máquinas de aprendizado extremo (ELM) - aqui denominadas coletivamente por máquinas desorganizadas (MDs), para a previsão de séries de vazões. A previsão de vazões é uma das etapas fundamentais no planejamento da operação dos sistemas de energia elétrica com predominância hidráulica, como é o caso brasileiro. Os modelos mais comumente utilizados para previsão de vazões pelo Setor Elétrico Brasileiro (SEB) são baseados na metodologia Box & Jenkins, lineares, sobretudo modelos periódicos auto-regressivos (PAR). Todavia, técnicas mais abrangentes, que alcancem melhores desempenhos, vêm sendo investigadas. Destacam-se as redes neurais artificiais, sobretudo arquiteturas do tipo perceptron de múltiplas camadas (MLP), muito conhecidas por serem aproximadores universais com elevada capacidade de aprendizado e mapeamento não-linear, características desejáveis para solução do problema em questão. Por outro lado, as máquinas desorganizadas têm apresentado resultados promissores na previsão de séries temporais. Estes modelos têm um processo de treinamento simples, baseado em encontrar os coeficientes de um combinador linear; em particular, não precisam fazer ajuste dos pesos de sua camada intermediária, ao contrário das redes MLP. Por isso, este trabalho investigou as MDs do tipo ESN e ELM, versões recorrente e não-recorrente, respectivamente, para previsão de vazões médias mensais. Serão avaliadas também três técnicas para retirada da componente sazonal característica destas séries ¿ médias móveis, padronização e diferenças sazonal ¿ além da exploração de técnicas de seleção de variáveis do tipo filtro e wrapper, no intuito de melhorar performance dos modelos preditores. Na maioria dos casos estudados, os resultados obtidos pelas MDs na previsão das séries associadas a importantes usinas hidrelétricas brasileiras - Furnas, Emborcação e Sobradinho - em cenários com horizontes variados, mostraram-se de melhor qualidade do que os obtidos pelo modelo PAR e as redes neurais MLPs / Abstract: This work explores the possibility of application of neural network architectures ¿ echo state networks (ESN) and extreme learning machines (ELM) ¿ collectively referred as unorganized machines (UMs), to seasonal streamflow series forecasting. Streamflow forecasting is one of the key steps in the planning of operation of power systems with hydraulic predominance, as in the Brazilian case. The models most commonly used to streamflow prediction by the Brazilian Electric Sector are based on the Box & Jenkins methodology, with linear and especially periodic autoregressive models. However, more extensive techniques that achieve better performances have been investigated to this task. We highlight artificial neural networks, especially architectures such as multilayer perceptron (MLP), known to be universal approximators with high learning ability skills ability to perform nonlinear mapping, desirable characteristics for the solution of this problem. On the other hand, unorganized machines have shown promising results in time series forecasting. These models have a simple training process, based on finding the coefficients of a linear combiner; they do not require adjustments in the weights of the hidden layer, which are necessary with MLP architecture. Therefore, this study investigated the UMs such as ESN and ELM, recurrent and nonrecurrent versions, respectively, to seasonal streamflow series forecasting. Three techniques to remove the seasonal component of streamflow series will also be evaluated - moving averages, standardization and seasonal differences. In addition, In order to improve the performance of predictive models techniques for variable selection, such as filters and wrappers, will also be explored. In the most cases, the computational results obtained by the UMs in streamflow series forecasting associated to important Brazilian hydroelectric plants - Furnas, Emborcação and Sobradinho - with scenarios including several horizons, presented better performance when compared to forecasting obtained with PAR models and MLPs / Doutorado / Energia Eletrica / Doutor em Engenharia Elétrica
|
32 |
Analýza sezónnosti v českém stavebnictví / Analysis of Seasonality in the Czech ConstructionŠimpach, Ondřej January 2010 (has links)
The output of the National economy of the Czech Republic is conditioned by a sum of important factors. There are sectors, which increased power during the last two decades, mainly due to expansion of modern technologies and knowledge workers. One of this is Construction. Construction is specific to its position in the economy and in particular is characterized by the greatest seasonality ever. However, this is not a problem for statistical analysis, rather a benefit. Modern approaches allow us to analyze seasonal fluctuations. From selected data we are able to construct evolutionary forecasts. The work will be performed for the most important indicators in the Czech Construction. The outcome of the paper will be conditional forecasts of these indicators. It will also make analyze of the relationship between these indicators and other variables that might affected it. The work is practical application of stochastic modeling approach by Box and Jenkins, augmented by more modern approaches, such as verification of Granger causality and co-integration and testing of seasonal unit roots by Hylleberg et al.
|
33 |
ANÁLISE COMPARATIVA DE MÉTODOS DE PREVISÃO DE SÉRIES TEMPORAIS ATRAVÉS DE MODELOS ESTATÍSTICOS E REDE NEURAL ARTIFICIAL. / COMPARATIVE ANALYSIS OF TIME SERIES FORECASTING METHODS THROUGH STATISTICAL MODELS AND ARTIFICIAL NEURAL NETWORKS.Sousa, Ana Paula de 09 March 2012 (has links)
Made available in DSpace on 2016-08-10T10:40:27Z (GMT). No. of bitstreams: 1
ANA PAULA DE SOUSA.pdf: 965882 bytes, checksum: a3647999f994441f4537855527b52292 (MD5)
Previous issue date: 2012-03-09 / The objective of this study was to compare statistical methods and artificial intelligence
to the problem of time series forecasting using Holt-Winters, Box-Jenkins and the
Elman neural network. The models were used to predict one step ahead of the price of
ethanol in the state of Goias and compared using measures of specific errors. At the end,
the results indicated that all three techniques were competitive in terms of predicting
one step ahead especially the statistical models appeared to be the most suitable
methods in terms of balance between performance and complexity. / O objetivo deste trabalho foi comparar os métodos de estatística e de inteligência
artificial para o problema da previsão de séries temporais através de Holt-Winters, Box-
Jenkins e a rede neural de Elman. Os modelos foram utilizados para previsão um passo
a frente dos preços do etanol no estado de Goiás e comparados através medidas de erros
específicas. Ao final, os resultados indicaram que todos os métodos se mostraram
competitivos em termos de predição um passo à frente, destacando-se os modelos
estatísticos como os mais adequados em termos de parcimônia entre desempenho e
complexidade.
|
34 |
Análise da difusão da emissão de certificações ISO 14001 nos países do continente americanoHIKICHI, Suzana Eda 25 February 2016 (has links)
O padrão ISO 14001 vem apresentando crescente importância para a gestão ambiental
nas organizações em todo o mundo. Estudar e prever o comportamento do número de
certificações nos próximos anos é uma estratégia para o planejamento e gestão organizacional.
O continente americano é uma importante região com relação à expansão da ISO 14001.
Entretanto poucos estudos relacionados à difusão da norma foram realizados com o foco nos
países do continente. Pelo exposto, o trabalho teve o objetivo de caracterizar a difusão da
norma ISO 14001 nos países e setores do continente americano e ajustar modelos de previsão
para o número de certificações. Em nível de países, o estudo foi realizado com os dados do
número de certificados ISO 14001 nos 13 países com o maior número de certificações entre
os anos de 1996 a 2014. A análise da ISO 14001 nos países foi realizada por meio das curvas
de evolução da ISO 14001. O comportamento da norma em nível de setores foi analisado
utilizando dados dos anos de 1999-2014. A relação entre o número de certificações ISO
14001 e ISO 9001 para países e setores foi avaliada por análise gráfica. No ajuste dos
modelos de previsão foi aplicada a metodologia de Box & Jenkins para as séries anuais de
dados dos 13 países. Os resultados da caracterização mostraram que a difusão vem ocorrendo
de forma heterogênea no continente. O interesse das organizações em certificar varia entre
países, dependendo de fatores internos. Em nível de setor, as certificações parecem estar
distribuídas entre diversos setores nos países. A disparidade do número de certificações ISO
14001 em relação à ISO 9001 está se reduzindo ao longo do tempo em países e nos setores.
Os modelos ARIMA ajustados para as séries ISO 14001 indicaram tendência de crescimento
no número de certificações no continente americano e no Brasil, Colômbia, Estados Unidos e
México. Chile e Argentina teriam tendência de redução no número de novos certificados.
Esses resultados propõem que os países que possuem condições de implantar a ISO 14001
devem permanecer interessados nos próximos anos. As informações obtidas nesse trabalho
podem ser úteis para orientar as estratégias de planejamento de organismos certificadores e a
gestão de negócios / The ISO 14001 standard has been showing increasing importance for environmental
management in organizations worldwide. To study and predict the behavior of the number of
certifications in coming years is a strategy for planning and organizational management. The
Americas is an important region in relation to the expansion of ISO 14001. However few
studies related to the dissemination of the standard were carried out with the focus on the
continent. Accordingly, this work aimed to characterize the diffusion of ISO 14001 standard
in the countries and sectors in the Americas and adjust predictive models for the number of
certifications. At the country level, the study was conducted with the number data of the ISO
14001 certificate in the 13 countries with the largest number of certifications between the
years of 1996 to 2014. The ISO 14001 analysis was carried out in the countries through the
evolution curves of the ISO 14001. The behavior of the standard at level of sectors was
analyzed using data from the years of 1999 to 2014. The relationship between the number of
ISO 14001 and ISO 9001 certifications for countries and sectors was evaluated by graphical
analysis. In adjusting of the forecast models were applied the Box & Jenkins methodology for
the annual series data from 13 countries. The results of the characterization showed that the
diffusion is heterogeneously on the continent. The interest of organizations to certify varies
between countries depending on internal factors. At the sector level, the certifications appear
to be distributed among various sectors in the countries. The disparity between the number of
ISO 14001 certifications in relation to ISO 9001 is being reduced over time in countries and
sectors. The ARIMA models adjusted to the ISO 14001 series showed a growth trend in the
number of certifications in the Americas and in Brazil, Colombia, United States and Mexico.
Chile and Argentina have downward trend in the number of new certificates. These results
suggest that countries that have conditions to implement the ISO 14001 should remain
interested in coming years. The information obtained in this work can be useful to guide the
planning strategies of certification bodies and business management. / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES
|
35 |
A model for the generation and study of electromyographic signalsLerman, David 05 December 1991 (has links)
A computer model simulating the electrical activity of muscles of the upper
arm during elbow motion is presented. The output of the model is an
Electromyographic (EMG) signal. System identification is performed on the EMG
signals using autoregressive moving average (ARMA) modelling. The calculated
ARMA coefficients are then used as the feature set for pattern recognition.
Pattern recognition is performed on the EMG signals to attempt to identify which
of four possible motions is producing the signal. The results of pattern recognition
are compared with results from pattern recognition of real EMG signals. The
model is shown to be useful in predicting general trends found in the real data, but
is not robust enough to predict accurate quantitative results. Simplifying
assumptions about the filtering effects of body tissue, and about the size and
position of muscles, are conjectured to be the most likely reasons the model is not
quantitatively accurate. / Graduation date: 1992
|
36 |
The application of Box-Jenkins models to the forecast of time series of Mainland China tourists in MacaoNgan, Wai Seng January 2011 (has links)
University of Macau / Faculty of Science and Technology / Department of Mathematics
|
37 |
Vliv makroekonomických faktorů na hodnotu indexu PXKohoutek, Jiří January 2015 (has links)
Kohoutek, J. Influence of macroeconomics factors on index PX. Diploma thesis. Brno: Mendel University, 2015. Aim of the thesis is theoretically and empirically explain the influence of macro-economic factors on the value of the index PX . For the analysis are used selected macroeconomic indicators and indexes of other exchanges. The time period of this data is from 2000 to 2013. The design is the least squares method and analysis of time series according to Box- Jenkins methodology . Part of this work is familiar with working on a similar theme other authors. In conclusion, results are evaluated and outlined the possibility of both the value of the index PX predict the future.
|
38 |
Métodos Box-Jenkins e neuro-nebuloso afins aplicados à previsão de séries temporais / Aline Purcote ; orientador, Leandro dos Santos CoelhoPurcote, Aline January 2009 (has links)
Dissertação (mestrado) - Pontifícia Universidade Católica do Paraná, Curitiba, 2009 / Bibliografia: f. 94-97 / O planejamento e abordagens de apoio à tomada de decisão podem ter a previsão como um dos seus subsídios, uma vez que fornece informações que possibilitam o planejamento com antecedência e, conseqüentemente, permite que os recursos produtivos estejam disp / The planning and approaches to support decision-making can be forecasting as one of its subsidies, as it provides information that enable the planning in advance and hence allows productive resources that are available in quantity, time and quality. Assis
|
39 |
Sistemas de previsão de preços de commodities no mercado futuroSantos, Jair Pereira dos 14 May 1993 (has links)
Made available in DSpace on 2010-04-20T20:08:12Z (GMT). No. of bitstreams: 0
Previous issue date: 1993-05-14T00:00:00Z / Este trabalho compara procedimentos de previsão de preços de commodities, utilizados de maneira empírica pelos analistas de mercado, com os procedimentos fornecidos pela Análise de Séries Temporais. Aplicamos os métodos de previsão utilizando as Médias Móveis, os métodos baseados em Alisamentos exponenciais e principalmente os modelos ARIMA de Box-Jenkins. Estes últimos são, em geral, generalizações dos primeiros, com a vantagem de utilizar os instrumentos estatísticos de medidas das incertezas, como o desvio-padrão e os intervalos de confiança para as previsões.
|
40 |
Obchodování s kávou na komoditních trzích / Coffee Trading on Commodity MarketsKašička, Jan January 2015 (has links)
This thesis offers a comprehensive view of coffee trading on commodity markets. To describe the behavior of prices and their volatility, ARCH and GARCH models are used. These models analyse coffee prices of selected regions in Ethiopia, the birth place of coffee. The thesis connects the characteristics of soft commodity with current knowledge of financial econometrics. It also describes the effects of changes in exchange rates and oil prices on the price of coffee. Price volatility is examined with regard to deregulation and reforms on this market within the last three decades. Developments in the developing world caused a significant need for the progression and identification with the hedging instruments. These are closely linked to the globalized market with coffee, so it is conversely possible to absorb the shocks on small growers, who are significantly impacted by the globalized world.
|
Page generated in 0.0192 seconds