In some situations we cannot observe the original time series and instead, we record only binary data which express whether the values of the original series exceeded a certain threshold or not. The thesis deals with estimation of characteristics of the original series constructed from the binary (so called clipped or hard-limited) data, in particular in Gaussian ARMA models. We summarize some basic characteristics of the clipped series and describe their relation to the original ones. Some practical examples are provided as well. The estimation of parameters in AR(p) model is shown for the case of zero threshold. Using a similar approach, an estimator of the MA(1) model parameter is proposed and its properties are studied with emphasis on asymptotic variance. Subsequently, we propose an estimation procedure for AR(p) and MA(1) models with unknown (non-zero) threshold. The behaviour of our estimators is investigated in a simulation study, which provides a comparison with estimators constructed from the original data. Finally, a real data analysis is presented for an illustration. Powered by TCPDF (www.tcpdf.org)

Disertaciniame darbe nagrinėjama erdvės – laiko duomenų modeliavimo problema. Dažnai erdvinių duomenų rinkiniai yra gana nedideli, o taškai, kuriuose pasklidę stebėjimai, išsidėstę netaisyklingai. Sprendžiant „erdvinį“ uždavinį, paprastai siekiama inerpoliuoti arba įvertinti erdvinį vidurkį. Laiko eilučių duomenys dažniausiai naudojami ateities reikšmėms prognozuoti. Tuo tarpu erdvės – laiko uždaviniai jungia abu uždavinių tipus. Pasiūlyta keletas originalių erdvinių laiko eilučių modeliavimo metodų. Siūlomi metodai pirmiausia analizuoja vienmates laiko eilutes, o pašalinus laikinę priklausomybė jose, laiko eilučių liekanoms vertinama erdvinė priklausomybė. Tikslas – sudaryti modelį, leidžiantį prognozuoti požymio reikšmę naujame, nestebėtame taške, nauju laiko momentu. Tokio modelio sudarymas remiasi laiko eilučių parametrų erdviniu interpoliavimu. / Space – time data modeling problem is analysed. Often spatial data sets are relatively small, and the points, where observations are taken, are located irregularly. When solving spatial task, usually we are interpolating or estimating the spatial average. Time series data usually are used to predict future values. Meanwhile, the space - time tasks combines both types of tasks. Few original modeling methods of spatial time series are proposed. The proposed methods firstly analyzes the univariate time series, and after removing temporal dependence, spatial dependence in the time series of residuals is measured. Aim of this dissertational work - to create time series model at new unobserved location by incorporating spatial interaction thru spatial interpolation of estimated time series parameters. Such a model is based on the spatial interpolation of time series parameters.

Space – time data modeling problem is analysed. Often spatial data sets are relatively small, and the points, where observations are taken, are located irregularly. When solving spatial task, usually we are interpolating or estimating the spatial average. Time series data usually are used to predict future values. Meanwhile, the space - time tasks combines both types of tasks. Few original modeling methods of spatial time series are proposed. The proposed methods firstly analyzes the univariate time series, and after removing temporal dependence, spatial dependence in the time series of residuals is measured. Aim of this dissertational work - to create time series model at new unobserved location by incorporating spatial interaction thru spatial interpolation of estimated time series parameters. Such a model is based on the spatial interpolation of time series parameters. / Disertaciniame darbe nagrinėjama erdvės – laiko duomenų modeliavimo problema. Dažnai erdvinių duomenų rinkiniai yra gana nedideli, o taškai, kuriuose pasklidę stebėjimai, išsidėstę netaisyklingai. Sprendžiant „erdvinį“ uždavinį, paprastai siekiama inerpoliuoti arba įvertinti erdvinį vidurkį. Laiko eilučių duomenys dažniausiai naudojami ateities reikšmėms prognozuoti. Tuo tarpu erdvės – laiko uždaviniai jungia abu uždavinių tipus. Pasiūlyta keletas originalių erdvinių laiko eilučių modeliavimo metodų. Siūlomi metodai pirmiausia analizuoja vienmates laiko eilutes, o pašalinus laikinę priklausomybė jose, laiko eilučių liekanoms vertinama erdvinė priklausomybė. Tikslas – sudaryti modelį, leidžiantį prognozuoti požymio reikšmę naujame, nestebėtame taške, nauju laiko momentu. Tokio modelio sudarymas remiasi laiko eilučių parametrų erdviniu interpoliavimu.

Various models, such as ARMA and GARCH, are used in the financial time series framework. The purpose of this thesis is to present an alternative for these models which are bilinear time series models. First chapter is theore- tical, there is a short introduction to the theory of time series and ARMA models. Second chapter focuses on theoretical aspects of the simple bilinear model, third chapter presents the theory for general bilinear model in the similiar fashion as for simple model. Last chapter is focused on practical aspects, it contains simulations and examines the properties of estimates based on the presented theory, final part is devoted to the comparison of properties of ARMA models and bilinear models for selected financial data. 1

[pt] O bootstrap de B. Efron, que não poderia ser imaginado sem os computadores de hoje, pode resolver vários problemas livre da suposição de Gaussianidade para os dados. Este trabalho tem o objetivo de apresentar essa técnica computacionalmente intensiva no contexto de Séries temporais - Metodologia Box and Jenkins. Como se sabe essa Metodologia possui alguns resultados assintóticos. Então, na fase da identificação da estrutura do modelo, pode apresentar problemas em regiões do espaço paramétrico aqui determinadas,. O bootstrap é proposto como opção e um estudo de simulação, comparativo, é apresentado. Constrói- se a distribuição bootstrap da autocorrelação e autocorrelação parcial, amostrais, e ainda a distribuição bootstrap do estimador de MQNL dos coeficientes de modelos ARMA (p, q). consequentemente, fica disponí­vel medida não- paramétrica da precisão da estimativa. O estudo de simulação que aborda o estimador de MQNL dos coeficientes enfoca, basicamente, a região de fronteira da estacionariedade e inversibilidade. / [en] The bootstrap of B. Efron, what should not be imagined without fast andcheaper computation, can solve several problems free from assumption that the data conform to a bell-shaped curve. This work has the aim to present this computer-intensive technics in the context of Time Series - Box and Jenkins´s Methodology. As we know this methodology own some asymptotic results. Then in the identification stage of the structure of the model it may present some troubles on regions of the parametric space, as we show later on the bootstrap is proposed as an aption and a comparative simulation study is pointed out. We build up the bootstrap distribution of the sample autocorrelation and sample partial autocorrelation, and yet a bootstrap distribution to the non-linear LS estimator of the coefficients to the ARMA (p,q) model. As a consequence we get the non- parametric measure of the accuracy of the estimates. The study of simulation wich takes into account the non-linear LS estimato to the coefficients, actually focalize the borden of the stationarity and invertibility region.

This thesis attempts to predict daily number of firefighter incidents in the Central Bohemia Region and in the Region of Hradec Králové to improve firefighter shift planning. The analysis is based on a dataset of firefighter incidents from the period between the years 2008 and 2012. Econometric models, capturing yearly and weekly patterns and weather impact were estimated and used for long-term prediction. The first part of the thesis provides a description of tests applied to residuals and other econometric tests used in this study. Then linear regression is applied to model weather impact and effects of days of week and months of year. In the next part regression with AR errors, (S)ARMA models and regression with (S)ARMA errors are estimated. All these models are compared according to properties of residuals and out-of-sample mean absolute percentage error (MAPE). The most accurate models predict daily number of incidents two months ahead with MAPE slightly above 20% which is considerably better than the benchmark Holt-Winters method. Regression models with (S)ARMA errors produce relatively accurate long-term forecasts and its error terms are uncorrelated. Therefore, they can be considered suitable for long-term prediction of firefighter incidents.

Made available in DSpace on 2016-01-26T12:51:16Z (GMT). No. of bitstreams: 1 marcusviviciusbaptista_dissert.pdf: 5989384 bytes, checksum: 709f4e156c11b3804c116e8d1679f733 (MD5) Previous issue date: 2007-07-31 / In Brazil, there was a significant increase of deaths due to external causes resulting from wounds of firearm projectiles. This kind of wound is the third cause of Brazilians? deaths, after cardiac and cerebral vascular diseases. Taking into account the necessity of public and social order, both the Medicine and Forensic Justice have interest in victims of wounds of external causes. In this case, Justice searches in the medical science a set of juridical knowledge to help it to defend the rights and interests of men and society: this is the Forensics Medicine. The Forensic Physician has to be able to provide the causa mortis as well to describe the wounds of the Justice interest. In the cases associated with firearms, this physician has to determine if the orifice from the projectile in the body is of entrance or exit. This is very important according to the juridical and proceeding point of view. The firearm projectiles cause in the human skin several typical wounds; however, the shot direction or if the wound occurred due to the entrance or the exit of the projectile are not always determined macroscopically by the forensic physician. Therefore, other means, as the histological, to provide an accurate diagnosis should be taken into account. This study analyzed macroscopically and microscopically 14 wounds of entrance and their 14 wounds of exits, comparing data with the related in the literature. This study showed that hystopathological analysis of wounds due to firearm projectiles provides an effective diagnosis of the entrance and exit orifice. Coagulation necrosis of keratinocytes associated with subepidermal fissure (effects of burn injuries) is a pathognomonic feature of the entrance orifice, whereas the presence of adipose, muscle and bone tissues in the derm is characteristic of the exit orifice. / No Brasil, nos ?ltimos anos, houve um aumento acentuado de mortes por causas externas, devido a ferimentos de proj?teis de arma de fogo. Atualmente, os ferimentos causados por armas de fogo s?o a terceira causa de morte entre os brasileiros, atr?s somente das doen?as card?acas e das c?rebro-vasculares. As v?timas dos ferimentos por causas externas s?o de interesse da Medicina e tamb?m da Justi?a, levando-se em considera??o as necessidades de ordem p?blica e social. Neste caso, a Justi?a busca na ci?ncia m?dica um conjunto de conhecimentos extrajur?dicos para auxili?-la a defender os direitos e os interesses dos homens e da sociedade: ? a Medicina Legal. Compete ao M?dico Legista estabelecer a causa mortis e descrever os ferimentos de interesse da Justi?a. Nos casos que se correlacionam com armas de fogo, compete ao mesmo determinar se o orif?cio decorrente do proj?til no corpo humano ? de entrada ou de sa?da, tendo este fato suma import?ncia do ponto de vista jur?dico e processual. Os proj?teis de arma de fogo ocasionam na pele humana uma s?rie de ferimentos t?picos, mas nem sempre possibilitam o m?dico legista estabelecer, do ponto de vista macrosc?pico, a dire??o do tiro ou se um ferimento foi ocasionado pela entrada do proj?til ou pela sua sa?da. Deve-se, ent?o, lan?ar m?os de meios mais precisos ? histol?gicos ? para se estabelecer o diagn?stico de certeza. Este estudo analisou 14 ferimentos de entrada e suas 14 respectivas sa?das, do ponto de vista macrosc?pico e microsc?pico, comparando-se com os dados j? existentes na literatura. Este estudo demonstrou que a an?lise histopat?logica dos ferimentos ocasionados por proj?teis de arma de fogo fornece substrato para o diagn?stico mais acurado do orif?cio de entrada e de sa?da, sendo que a necrose de coagula??o dos ceratin?citos associada ? fenda sub-epid?rmica (efeitos de les?o por queimadura) ? caracter?stica de orif?cios de entrada e a presen?a de tecido adiposo, tecido muscular e tecido ?sseo, na derme, ? caracter?stica de orif?cio de sa?da.

Submitted by Aelson Maciera (aelsoncm@terra.com.br) on 2017-08-03T20:04:27Z No. of bitstreams: 1 TeseBSA.pdf: 10322083 bytes, checksum: 4c30c490934f23dbad9d5a1f087ef182 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-08-08T19:09:23Z (GMT) No. of bitstreams: 1 TeseBSA.pdf: 10322083 bytes, checksum: 4c30c490934f23dbad9d5a1f087ef182 (MD5) / Approved for entry into archive by Ronildo Prado (ronisp@ufscar.br) on 2017-08-08T19:09:30Z (GMT) No. of bitstreams: 1 TeseBSA.pdf: 10322083 bytes, checksum: 4c30c490934f23dbad9d5a1f087ef182 (MD5) / Made available in DSpace on 2017-08-08T19:15:39Z (GMT). No. of bitstreams: 1 TeseBSA.pdf: 10322083 bytes, checksum: 4c30c490934f23dbad9d5a1f087ef182 (MD5) Previous issue date: 2016-12-16 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Generalized autoregressive moving average (GARMA) models are a class of models that was developed for extending the univariate Gaussian ARMA time series model to a flexible observation-driven model for non-Gaussian time series data. This work presents the GARMA model with discrete distributions and application of resampling techniques to this class of models. We also proposed The Bayesian approach on GARMA models. The TGARMA (Transformed Generalized Autoregressive Moving Average) models was proposed, using the Box-Cox power transformation. Last but not least we proposed the Bayesian approach for the TGARMA (Transformed Generalized Autoregressive Moving Average). / Modelos Autoregressivos e de médias móveis generalizados (GARMA) são uma classe de modelos que foi desenvolvida para extender os conhecidos modelos ARMA com distribuição Gaussiana para um cenário de series temporais não Gaussianas. Este trabalho apresenta os modelos GARMA aplicados a distribuições discretas, e alguns métodos de reamostragem aplicados neste contexto. É proposto neste trabalho uma abordagem Bayesiana para os modelos GARMA. O trabalho da continuidade apresentando os modelos GARMA transformados, utilizando a transformação de Box-Cox. E por último porém não menos importante uma abordagem Bayesiana para os modelos GARMA transformados.

Conventional time series analysis depends heavily on the twin assumptions of linearity and stationarity. However; there are certain cases where sampled data tend to violate the assumptions. In this paper, we use neural networks technology to explore the situation when the assumptions of linearity and stationarity are failed. At the end of the paper, we discuss an illustrative example about the annual expenditures of government and science-education-culture of R.O.C.

L'objet de la thèse est d'exposer une démarche méthodologique qui vise à prendre en compte, dans les modèles de séries temporelles, des effets exogènes mesurés à l'aide de variables additionnelles, et de l'illustrer par un certain nombre d'applications au secteur des transports. Dans ces applications, le pas de temps est le jour, le mois, le trimestre voire le semestre : il s'est agi de prendre en compte des effets exogènes, de nature transitoire ou de nature durable, qui se manifestent dans le court terme. La première partie de la thèse traite de la modélisation des séries temporelles. Nous situons le cadre formel des modèles auxquels nous nous intéressons, nous exposons la démarche suivie dans le cadre des modèles ARMA avec variables explicatives, puis dans le cadre des modèles markoviens avec variables explicatives en y détaillant le cas particulier des modèles structurels. Les deuxième et troisième parties de la thèse regroupent deux ensembles d'applications. Le premier porte sur des données de trafics, de voyageurs et de marchandises, agrégées par mode de transport ou par grande catégorie de réseau, et le second sur des données d'accidents corporels et de victimes de la circulation routière, agrégées par grande catégorie de réseau routier. La période couverte la plus large est 1970-2000. La plupart des applications intègre la prise en compte des effets transitoires, de nature climatique et calendaire, sur la demande de transport et sur le risque routier, et nous donnons dans la thèse les premiers résultats détaillés démontrant pour la France la significativité du facteur climatique sur le bilan routier national, mesuré en nombres d'accidents corporels et de tués