• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • 1
  • Tagged with
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

An application of exponential smoothing methods to weather related data

Marera, Double-Hugh Sid-vicious January 2016 (has links)
A Research Report submitted to the Faculty of Science in partial fulfilment of the requirements for the degree of Master of Science in the School of Statistics and Actuarial Science. 26 May 2016 / Exponential smoothing is a recursive time series technique whereby forecasts are updated for each new incoming data values. The technique has been widely used in forecasting, particularly in business and inventory modelling. Up until the early 2000s, exponential smoothing methods were often criticized by statisticians for lacking an objective statistical basis for model selection and modelling errors. Despite this, exponential smoothing methods appealed to forecasters due to their forecasting performance and relative ease of use. In this research report, we apply three commonly used exponential smoothing methods to two datasets which exhibit both trend and seasonality. We apply the method directly on the data without de-seasonalizing the data first. We also apply a seasonal naive method for benchmarking the performance of exponential smoothing methods. We compare both in-sample and out-of-sample forecasting performance of the methods. The performance of the methods is assessed using forecast accuracy measures. Results show that the Holt-Winters exponential smoothing method with additive seasonality performed best for forecasting monthly rainfall data. The simple exponential smoothing method outperformed the Holt’s and Holt-Winters methods for forecasting daily temperature data.
2

Remigração na profundidade mediante a equação da onda imagem / Depth remigration by means of the image wave equation

Munerato, Fernando Perin 31 March 2006 (has links)
Orientadores: Joerg Schleicher, Amelia Novais / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica / Made available in DSpace on 2018-08-06T03:30:50Z (GMT). No. of bitstreams: 1 Munerato_FernandoPerin_M.pdf: 1553622 bytes, checksum: 5cb80dbf31a93da9d201b78855292dfc (MD5) Previous issue date: 2006 / Resumo: Este trabalho aborda a questão de como resolver a equação da onda imagem para o problema de remigração na profundidade através de métodos numéricos. O objetivo deste problema é a reconstrução de uma imagem das camadas geológicas do subsolo a partir de uma imagem previamente migrada com um modelo de velocidade, geralmente, incorreto. Nosso principal objetivo neste trabalho é a investigação de possíveis métodos que possam resolver os problemas que surgiram ao usarmos esquemas explícitos do método de diferenças _nitas na solução da equação da onda imagem em trabalhos anteriores, como, por exemplo, a dispersão numérica. Para isso, estudamos aqui o método de volumes _nitos, assim como esquemas implícitos do método de diferenças _nitas. O método de volumes _nitos possui como característica principal propagar as médias das células da malha ao invés de simplesmente os dados pontuais como é feito no método de diferenças _nitas. As outras tentativas para solucionar o problema da dispersão foram dois tipos de implementação de esquemas implícitos do método de diferenças _nitas, isto é, implementações implícitas de esquemas convencionais avaliados em pontos da malha e um esquema avaliado nos centros das células. A qualidade dos algoritmos estudados foi testada numericamente. Estes testes numéricos mostram que o método de volumes _nitos não é adequado para resolver o problema da dispersão, uma vez que a média calculada a cada passo aumenta o estiramento do pulso. Além disso, as implementações implícitas dos esquemas convencionais mostram o mesmo comportamento de dispersão que as implementações explícitas. Unicamente o esquema centrado foi capaz de melhorar a dispersão numérica em comparação com as implementações anteriores,porém somente para dados contendo exclusivamente baixas freqüências / Abstract: This work approaches the question of how to solve the image-wave equation for depth remigration by numerical methods. The objective is the reconstruction of an image of the geologic layers of the subsoil from a previously migrated image with a different velocity model. Our main objective in this work is the investigation of possible methods that can solve the problems that appeared when using explicit _nite-difference schemes for the solution of the image-wave equation in previous works, particularly numerical dispersion. For this purpose, we study the method of _nite volumes, as well as implicit _nite-difference schemes. The main characteristic of the _nite-volume method is to simply propagate the averages in the cells of the mesh instead of the discretized data themselves as it is done in the _nitedifference method. As another attempt to solve the problem of the dispersion, we study two types of implementation of implicit _nite-difference schemes, that is, implicit implementations of conventional schemes evaluated out the edge of the cell and a scheme evaluated in the center of the cell. The quality of the studied algoritms has been tested numerically. These numerical tests show that the method of _nite volumes is not adequate to solve the problem of dispersion, for the average calculated in each step additionally increases the pulse stretch. Moreover, the implicit implementations of the conventional schemes show the same dispersion behavior as the explicit implementations. Solely the centered scheme was capable to improve the numerical dispersion in comparison with the previous implementations, however only for data containing / Mestrado / Geofisica Computacional / Mestre em Matemática Aplicada

Page generated in 0.0477 seconds