Dimension reduction is a vital tool in many areas of applied statistics in which the dimensionality of the predictors can be large. In such cases, many statistical methods will fail or yield unsatisfactory results. However, many data sets of high dimensionality actually contain a much simpler, low-dimensional structure. Classical methods such as principal components analysis are able to detect linear structures very effectively, but fail in the presence of nonlinear structures. In the first part of this thesis, we investigate the asymptotic behavior of two nonlinear dimensionality reduction algorithms, LTSA and HLLE. In particular, we show that both algorithms, under suitable conditions, asymptotically recover the true generating coordinates up to an isometry. We also discuss the relative merits of the two algorithms, and the effects of the underlying probability distributions of the coordinates on their performance. Model selection is a fundamental problem in nearly all areas of applied statistics. In particular, a balance must be achieved between good in-sample performance and out-of-sample prediction. It is typically very easy to achieve good fit in the sample data, but empirically we often find that such models will generalize poorly. In the second part of the thesis, we propose a new procedure for the model selection problem which generalizes traditional methods. Our algorithm allows the combination of existing model selection criteria via a ranking procedure, leading to the creation of new criteria which are able to combine measures of in-sample fit and out-of-sample prediction performance into a single value. We then propose an algorithm which provably finds the optimal combination with a specified probability. We demonstrate through simulations that these new combined criteria can be substantially more powerful than any individual criterion.

No description available.

Potentially valuable information about the underlying data generating process of a dependent variable is often lost when an independent variable is transformed to fit into the same sampling frequency as a dependent variable. With the mixed data sampling (MIDAS) technique and increasingly available data at high frequencies, the issue of choosing an optimal sampling frequency becomes apparent. We use financial data and the MIDAS technique to estimate thousands of regressions and forecasts in the quarterly, monthly, weekly, and daily sampling frequencies. Model fit and forecast performance measurements are calculated from each estimation and used to generate summary statistics for each sampling frequency so that comparisons can be made between frequencies. Our regression models contain an autoregressive component and five additional independent variables and are estimated with varying lag length specifications that incrementally increase up to five years of lags. Each regression is used to forecast a rolling, one and two-step ahead, static forecast of the quarterly Yen and U.S Dollar spot exchange rate. Our results suggest that it may be favourable to include high frequency variables for closer modeling of the underlying data generating process but not necessarily for increased forecasting performance. / Graduate / 0501 / 0508 / 0511 / benjamincann@gmail.com

Submitted by Franciele Moreira (francielemoreyra@gmail.com) on 2018-03-29T12:30:01Z No. of bitstreams: 2 Dissertação - Marcelo dos Santos Ventura - 2018.pdf: 4739813 bytes, checksum: 52211670f6e17c893ffd08843056f075 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2018-03-29T13:40:08Z (GMT) No. of bitstreams: 2 Dissertação - Marcelo dos Santos Ventura - 2018.pdf: 4739813 bytes, checksum: 52211670f6e17c893ffd08843056f075 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2018-03-29T13:40:08Z (GMT). No. of bitstreams: 2 Dissertação - Marcelo dos Santos Ventura - 2018.pdf: 4739813 bytes, checksum: 52211670f6e17c893ffd08843056f075 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2018-03-09 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / This work deals with two Monte Carlo simulation studies in log-symmetric regression models, which are particularly useful for the cases when the response variable is continuous, strictly positive and asymmetric, with the possibility of the existence of atypical observations. In log- symmetric regression models, the distribution of the random errors multiplicative belongs to the log-symmetric class, which encompasses log-normal, log- Student-t, log-power- exponential, log-slash, log-hyperbolic distributions, among others. The first simulation study has as objective to examine the performance for the maximum-likelihood estimators of the model parameters, where various scenarios are considered. The objective of the second simulation study is to investigate the accuracy of popular information criteria as AIC, BIC, HQIC and their respective corrected versions. As illustration, a movie data set obtained and assembled for this dissertation is analyzed to compare log-symmetric models with the normal linear model and to obtain the best model by using the mentioned information criteria. / Este trabalho aborda dois estudos de simulação de Monte Carlo em modelos de regressão log- simétricos, os quais são particularmente úteis para os casos em que a variável resposta é contínua, estritamente positiva e assimétrica, com possibilidade da existência de observações atípicas. Nos modelos de regressão log-simétricos, a distribuição dos erros aleatórios multiplicativos pertence à classe log-simétrica, a qual engloba as distribuições log-normal, log-Student- t, log-exponencial- potência, log-slash, log-hyperbólica, entre outras. O primeiro estudo de simulação tem como objetivo examinar o desempenho dos estimadores de máxima verossimilhança desses modelos, onde vários cenários são considerados. No segundo estudo de simulação o objetivo é investigar a eficácia critérios de informação populares como AIC, BIC, HQIC e suas respectivas versões corrigidas. Como ilustração, um conjunto de dados de filmes obtido e montado para essa dissertação é analisado para comparar os modelos de regressão log-simétricos com o modelo linear normal e para obter o melhor modelo utilizando os critérios de informação mencionados.

Time series modeling and forecasting has many applicability in scientific and technological researchs. Specifically about variables restricted to the interval (0; 1), which includes rates and proportions, the classical regression models could not be suitable because they assume normality. In this context, Rocha and Cribari-Neto (2009) proposed the beta autoregressive moving average (βARMA) model. It admits that the variable of interest is beta distributed. The beta distribution is more flexible than the normal distribution and also assumes that de dependent variable is restricted to the interval (0; 1). Through βARMA is possible to obtain results closer to the nature of the data. But just choose the better parametric model does not guarantee the accuracy of the fitted model. To identify the lags is also relevant to ensure the accuracy of the adjusted model. It is in this purpose that the model selection criteria, or information criteria, were developed. They compare the explanatory capacity of a group of models and select, among this group, the model which minimizes the information loss. In this context, this paper aims to evaluate by Monte Carlo simulations the performance of different selection criteria in βARMA model. Considering several scenarios and sample sizes, the selection criteria evaluated was AIC, BIC, HQ, AICc, BICc and HQc. The results indicate that BICc, HQ and HQc had the better performance identifying the true model among the candidate models. Using the selection criteria indicated by the simulation study, were also adjusted βARMA models to real data. It were considered the credit delinquency and the relationship between payroll loan and individual credit, both variables are from national financial system. It was adjusted the classical ARIMA models too. This models were compared with βARMA in applications. For both variables was found a reasonable proximity between the original data and the predicted by the models, with advantage for βARMA, as much inside as outside the sample. / A modelagem e a previsão de séries temporais é um campo de ampla aplicabilidade em diversas áreas científicas e tecnológicas. No âmbito específico de variáveis restritas ao intervalo (0; 1), como taxas e proporções, a utilização de modelos clássicos, que supõem normalidade da variável de interesse, pode não ser adequada. Neste contexto, Rocha e Cribari-Neto (2009) propuseram o modelo beta autorregressivo de médias móveis (β ARMA). Por assumir que a variável de interesse possui distribuição beta, que é uma distribuição mais flexível que a normal e com suporte restrito ao intervalo (0; 1), o βARMA possibilita modelagens e previsões mais condizentes com a natureza desses dados. Contudo, apenas a escolha do modelo paramétrico mais adequado não garante a acurácia do modelo ajustado. A identificação das defasagens a serem incluídas também exerce um papel de relevância neste sentido. É neste propósito que foram desenvolvidos os critérios de seleção de modelos, ou critérios de informação. Estes comparam as capacidades de explicação entre um grupo de modelos candidatos e selecionam, dentro deste grupo, o modelo que minimiza a perda de informações. Diante do exposto, este trabalho tem o objetivo de avaliar, via simulações de Monte Carlo, o desempenho de diferentes critérios de seleção no modelo βARMA. Por meio de um extenso estudo de simulação, considerando diversos cenários e tamanhos amostrais, foram avaliados os desempenhos em amostras de tamanho finito dos critérios AIC, BIC, HQ, AICc, BICc e HQc. Como resultados numéricos gerais, destaca-se que os critérios HQ, BICc e HQc foram os que alcançaram os melhores níveis de identificação do modelo verdadeiro. Utilizando os critérios de seleção sugeridos no estudo de simulação também foram ajustados modelos βARMA a dados reais. Para isso, foram considerados o índice de inadimplência de crédito e a relação entre o crédito consignado e o crédito total pessoa física, ambos do Sistema Financeiro Nacional. Também foram ajustados os clássicos modelos ARIMA comparativamente ao modelo βARMA na realização de previsões e posterior comparação entre os resultados de ambas as aplicações. Para as duas variáveis há um grau razoável de proximidade entre os dados originais e previstos, com superioridade do βARMA tanto dentro quanto fora do conjunto de observações utilizado para estimação dos modelos.