The thesis contains the four chapters: Testing parameter constancy in unit root autoregressive models against continuous change; Dickey-Fuller type of tests against nonlinear dynamic models; Inference for unit roots in a panel smooth transition autoregressive model where the time dimension is fixed; Testing unit roots in nonlinear dynamic heterogeneous panels. In Chapter  1 we derive tests for parameter constancy when the data generating process is non-stationary against the hypothesis that the parameters of the model change smoothly over time. To obtain the asymptotic distributions of the tests we generalize many theoretical results, as well as new are introduced, in the area of unit roots . The results are derived under the assumption that the error term is a strong mixing. Small sample properties of the tests are investigated, and in particular, the power performances are satisfactory. In Chapter 2 we introduce several test statistics of testing the null hypotheses of a random walk (with or without drift) against models that accommodate a smooth nonlinear shift in the level, the dynamic structure, and the trend. We derive analytical limiting distributions for all tests. Finite sample properties are examined. The performance of the tests is compared to that of the classical unit root tests by Dickey-Fuller and Phillips and Perron, and is found to be superior in terms of power. In Chapter 3 we derive a unit root test against a Panel Logistic Smooth Transition Autoregressive (PLSTAR). The analysis is concentrated on the case where the time dimension is fixed and the cross section dimension tends to infinity. Under the null hypothesis of a unit root, we show that the LSDV estimator of the autoregressive parameter in the linear component of the model is inconsistent due to the inclusion of fixed effects. The test statistic, adjusted for the inconsistency, has an asymptotic normal distribution whose first two moments are calculated analytically. To complete the analysis, finite sample properties of the test are examined. We highlight scenarios under which the traditional panel unit root tests by Harris and Tzavalis have inferior or reasonable power compared to our test. In Chapter 4 we present a unit root test against a non-linear dynamic heterogeneous panel with each country modelled as an LSTAR model. All parameters are viewed as country specific. We allow for serially correlated residuals over time and heterogeneous variance among countries. The test is derived under three special cases: (i) the number of countries and observations over time are fixed, (ii) observations over time are fixed and the number of countries tend to infinity, and (iii) first letting the number of observations over time tend to infinity and thereafter the number of countries. Small sample properties of the test  show modest size distortions and satisfactory power being superior to the Im, Pesaran and Shin t-type of test. We also show clear improvements in power compared to a univariate unit root test allowing for non-linearities under the alternative hypothesis. / Diss. Stockholm : Handelshögskolan, 2004

[pt] A seleção de variáveis em modelos estatísticos é um problema importante, para o qual diferentes soluções foram propostas. Tradicionalmente, pode-se escolher o conjunto de variáveis explicativas usando critérios de informação ou informação à priori, mas o número total de modelos a serem estimados cresce exponencialmente a medida que o número de variáveis candidatas aumenta. Um problema adicional é a presença de mais variáveis candidatas que observações. Nesta tese nós estudamos diversos aspectos do problema de seleção de variáveis. No Capítulo 2, comparamos duas metodologias para regressão linear: Autometrics, que é uma abordagem geral para específico (GETS) baseada em testes estatísticos, e LASSO, um método de regularização. Diferentes cenários foram contemplados para a comparação no experimento de simulação, variando o tamanho da amostra, o número de variáveis relevantes e o número de variáveis candidatas. Em uma aplicação a dados reais, os métodos foram comparados para a previsão do PIB dos EUA. No Capítulo 3, introduzimos uma metodologia para seleção de variáveis em modelos regressivos e autoregressivos de transição suave (STR e STAR) baseada na regularização do LASSO. Apresentamos uma abordagem direta e uma escalonada (stepwise). Ambos os métodos foram testados com exercícios de simulação exaustivos e uma aplicação a dados genéticos. Finalmente, no Capítulo 4, propomos um critério de mínimos quadrados penalizado baseado na penalidade l1 do LASSO e no CVaR (Conditional Value at Risk) dos erros da regressão out-of-sample. Este é um problema de otimização quadrática resolvido pelo método de pontos interiores. Em um estudo de simulação usando modelos de regressão linear, mostra-se que o método proposto apresenta performance superior a do LASSO quando os dados são contaminados por outliers, mostrando ser um método robusto de estimação e seleção de variáveis. / [en] Variable selection in statistical models is an important problem, for which many different solutions have been proposed. Traditionally, one can choose the set of explanatory variables using information criteria or prior information, but the total number of models to evaluate increases exponentially as the number of candidate variables increases. One additional problem is the presence of more candidate variables than observations. In this thesis we study several aspects of the variable selection problem. First, we compare two procedures for linear regression: Autometrics, which is a general-to-specific (GETS) approach based on statistical tests, and LASSO, a shrinkage method. Different scenarios were contemplated for the comparison in a simulation experiment, varying the sample size, the number of relevant variables and the number of candidate variables. In a real data application, we compare the methods for GDP forecasting. In a second part, we introduce a variable selection methodology for smooth transition regressive (STR) and autoregressive (STAR) models based on LASSO regularization. We present a direct and a stepwise approach. Both methods are tested with extensive simulation exercises and an application to genetic data. Finally, we introduce a penalized least square criterion based on the LASSO l1- penalty and the CVaR (Conditional Value at Risk) of the out-of-sample regression errors. This is a quadratic optimization problem solved by interior point methods. In a simulation study in a linear regression framework, we show that the proposed method outperforms the LASSO when the data is contaminated by outliers, showing to be a robust method of estimation and variable selection.