A structural approach for modelling a statistical problem permits to introduce a contextual theory based in previous knowledge. This approach makes the parameters completely meaningful; but, in the intermediate steps, some unobservable characteristics are introduced because of their contextual meaning. When the model is completely specified, the marginalisation into the observed variables is operated in order to obtain a tatistical model. The variables can be discrete or continuous both at the level of unobserved and at the level of observed or manifest variables. We are sometimes faced, especially in behavioural sciences, with ordinal variables; this is the case of the so-called Likert scales. Therefore, an ordinal variable could be nterpreted as a discrete version of a latent concept (the discretization model). The normality of the latent variables simplifies the study of this model into the analysis of the structure of the covariance matrix of the "ideally" measured variables, but only a sub-parameter of these matrix can be identified and consistently estimated (i.e. the matrix of polychoric correlations). Consequently, two questions rise here: Is the normality of the latent variables testable? If not, what is the aspect of this hypothesis which could be testable?. In the discretization model, we observe a loss of information with related to the information contained in the latent variables. In order to treat this situation we introduce the concept of partial observability through a (non bijective) measurable function of the latent variable. We explore this definition and verify that other models can be adjusted to this concept. The definition of partial observability permits us to distinguish between two cases depending on whether the involved function is or not depending on a Euclidean parameter. Once the partial observability is introduced, we expose a set of conditions for building a specification test at the level of latent variables. The test is built using the encompassing principle in a Bayesian framework. More precisely, the problem treated in this thesis is: How to test, in a Bayesian framework, the multivariate normality of a latent vector when only a discretized version of that vector is observed. More generally, the problem can be extended to (or re-paraphrased in): How to test, in Bayesian framework, a parametric specification on latent variables against a nonparametric alternative when only a partial observation of these latent variables is available.

Over the last decade, U.S. crop prices have become significantly more volatile. Volatile markets pose increased risks for the agricultural market participants and create a need for reliable price forecasts. Research discussed in this paper aims to find different approaches to forecast crop cash prices based on the prices of related futures contracts. Corn, soybeans, soft red winter wheat, and cotton are the focus of this research. Since price data for these commodities is non-stationary, this paper used two approaches to solve this problem. The first approach is to forecast the difference in prices between current and future period and the second is to use the regimes. This paper considers the five-year moving average approach as the benchmark when comparing these approaches. This research evaluated model performance using R-squared, mean errors, root mean squared errors, the modified Diebold-Mariano test, and the encompassing test. The results show that both the difference model and the regime model render better performance than the benchmark in most cases, but without a significant difference between each other. Based on these findings, the regime model was used to make forecasts of the cash prices of corn and soybeans, the difference model was used to make predictions for cotton, and the benchmark was used to forecast the SRW cash price. / Master of Science / This research attempts to develop models to forecast cash prices of corn, soybeans, wheat and cotton using the underlying futures prices. Two alternative approaches are proposed. The difference model focuses on forecasting the differences between current and future time prices. The regime model uses external data to determine potential structural breaks in price relationships. The out-of-sample performance of these models is compared to the benchmark of a five-year average using various performance criteria. The results show that the regime model performs better for corn and soybeans, while the difference model is the best one for cotton. For wheat, the results are mixed, but the benchmark seems to show better performance than the proposed models.

Submitted by Alexis Maka (alexis.maka@ipea.gov.br) on 2014-01-07T17:09:06Z No. of bitstreams: 1 Dissertation - Alexis Maka.pdf: 4105460 bytes, checksum: 61a014ecaca774cf3ae64ddbdd8ea527 (MD5) / Approved for entry into archive by ÁUREA CORRÊA DA FONSECA CORRÊA DA FONSECA (aurea.fonseca@fgv.br) on 2014-01-28T18:27:24Z (GMT) No. of bitstreams: 1 Dissertation - Alexis Maka.pdf: 4105460 bytes, checksum: 61a014ecaca774cf3ae64ddbdd8ea527 (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2014-02-03T15:53:23Z (GMT) No. of bitstreams: 1 Dissertation - Alexis Maka.pdf: 4105460 bytes, checksum: 61a014ecaca774cf3ae64ddbdd8ea527 (MD5) / Made available in DSpace on 2014-02-03T15:53:53Z (GMT). No. of bitstreams: 1 Dissertation - Alexis Maka.pdf: 4105460 bytes, checksum: 61a014ecaca774cf3ae64ddbdd8ea527 (MD5) Previous issue date: 2013-11-27 / Esta tese é composta por três ensaios sobre testes empíricos de curvas de Phillips, curvas IS e a interação entre as políticas fiscal e monetária. O primeiro ensaio ('Curvas de Phillips: um Teste Abrangente') testa curvas de Phillips usando uma especificação autoregressiva de defasagem distribuída (ADL) que abrange a curva de Phillips Aceleracionista (APC), a curva de Phillips Novo Keynesiana (NKPC), a curva de Phillips Híbrida (HPC) e a curva de Phillips de Informação Rígida (SIPC). Utilizamos dados dos Estados Unidos (1985Q1--2007Q4) e do Brasil (1996Q1--2012Q2), usando o hiato do produto e alternativamente o custo marginal real como medida de pressão inflacionária. A evidência empírica rejeita as restrições decorrentes da NKPC, da HPC e da SIPC, mas não rejeita aquelas da APC. O segundo ensaio ('Curvas IS: um Teste Abrangente') testa curvas IS usando uma especificação ADL que abrange a curva IS Keynesiana tradicional (KISC), a curva IS Novo Keynesiana (NKISC) e a curva IS Híbrida (HISC). Utilizamos dados dos Estados Unidos (1985Q1--2007Q4) e do Brasil (1996Q1--2012Q2). A evidência empírica rejeita as restrições decorrentes da NKISC e da HISC, mas não rejeita aquelas da KISC. O terceiro ensaio ('Os Efeitos da Política Fiscal e suas Interações com a Política Monetária') analisa os efeitos de choques na política fiscal sobre a dinâmica da economia e a interação entre as políticas fiscal e monetária usando modelos SVARs. Testamos a Teoria Fiscal do Nível de Preços para o Brasil analisando a resposta do passivo do setor público a choques no superávit primário. Para a identificação híbrida, encontramos que não é possível distinguir empiricamente entre os regimes Ricardiano (Dominância Monetária) e não-Ricardiano (Dominância Fiscal). Entretanto, utilizando a identificação de restrições de sinais, existe evidência que o governo seguiu um regime Ricardiano (Dominância Monetária) de janeiro de 2000 a junho de 2008. / This dissertation consists of three essays on empirical testing of Phillips curves, IS curves, and the interaction between fiscal and monetary policies. The first essay ('Phillips Curves: An Encompassing Test') tests Phillips curves using an autoregressive distributed lag (ADL) specification that encompasses the accelerationist Phillips curve (APC), the New Keynesian Phillips curve (NKPC), the Hybrid Phillips curve (HPC), and the Sticky-Information Phillips curve (SIPC). We use data from the United States (1985Q1--2007Q4) and from Brazil (1996Q1--2012Q2), using the output gap and alternatively the real marginal cost as measure of inflationary pressure. The empirical evidence rejects the restrictions implied by the NKPC, the HPC, and SIPC, but does not reject those implied by the APC. The second essay ('IS Curves: An Encompassing Test') tests IS curves using an ADL specification that encompasses the traditional Keynesian IS curve (KISC), the New Keynesian IS curve (NKISC), and the Hybrid IS curve (HISC). We use data from the United States (1985Q1--2007Q4) and from Brazil (1996Q1--2012Q2). The evidence rejects the restrictions implied by the NKISC and the HISC, but does not reject those of the KISC. The third essay ('The Effects of Fiscal Policy and its Interactions with Monetary Policy in Brazil') analyzes the effects of fiscal policy shocks on the dynamics of the economy and the interaction between fiscal and monetary policy using structural vector autoregressions (SVARs). We test the Fiscal Theory of the Price Level for Brazil, analyzing the response of public sector liabilities to primary surplus shocks. For the hybrid identification we find that it is not possible to distinguish empirically between Ricardian (Monetary Dominance) and non-Ricardian (Fiscal Dominance) regimes. However, using sign restrictions there is some evidence that the government followed a Ricardian (Monetary Dominance) regime from January 2000 to June 2008.