nÃo hà / Este estudo investiga a hipÃtese de eficiÃncia de mercado, a qual designa que estratÃgias de previsibilidade baseadas no comportamento passado das sÃries de retornos de aÃÃes nÃo implicam a obtenÃÃo de lucros econÃmicos. SÃo analisados dados de 25 mercados, estendendo-se de janeiro de 1990 a janeiro de 2010. A metodologia principal consiste na aplicaÃÃo de cinco testes de raiz unitÃria para painel, entre os quais se destaca o de Pesaran, Smith e Yamagata (2009), o qual assume que existe um determinado nÃmero de variÃveis que sÃo simultaneamente afetadas por um dado conjunto de fatores comuns nÃo observados. Os resultados modificam-se conforme altera-se o poder dos testes. O principal teste aplicado, particularmente, rejeita a hipÃtese em questÃo, sinalizando a possibilidade de exploraÃÃo de certas ineficiÃncias para a obtenÃÃo de lucros adicionais. / This paper investigates the efficient market hypothesis, which indicates a situation where investors are not able to develop a familiarity with past patterns of returns in order to obtain extra profits. It is used a sample containing 25 markets over the period January 1990 to January 2010. Econometric Methodology consists in exploiting five unit root tests, between which Pesaran, Smith e Yamagata (2009) is in relief, which assumes that there exists a number of variables that are simultaneously affected by a given set of unobserved common factors. Main results reject the efficient market hypothesis, indicating possibilities of exploiting inefficiency for obtaining extra profits.

Panel data analysis is the modern approach of statistical and econometric modeling. The aim of the thesis is to estimate the gravity model of international trade in alcoholic beverages in the form of bilateral trade flow depending on the gross domestic product and other associated variables that facilitate trading. The data have a panel structure. Based on the results of panel unit root tests the stationarity of variables in the panel and the expected long-term relationship between the analyzed variables are tested. Gravity model is assuming the existence of long-term relationships built through various methods, such as pooling OLS estimate, fixed and random effects models, cointegrated regression DOLS and FMOLS. Cointegration relationship is verified by Pedroni panel test.

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal. / In this thesis, we deal with three particular issues in the literature on nonstationary time series. The first essay deals with various unit root tests in the context of structural change. The second paper studies some residual based tests in order to identify cointegration. Finally, in the third essay, we analyze several tests in order to identify additive outliers in nonstationary time series. The first paper analyzes the hypothesis that some time series can be characterized as stationary with a broken trend. We extend the class of M-tests and ADF test for a unit root to the case where a change in the trend function is allowed to occur at an unknown time. These tests (MGLS, ADFGLS) adopt the Generalized Least Squares (GLS) detrending approach to eliminate the set of deterministic components present in the model. We consider two models in the context of the structural change literature. The first model allows for a change in slope and the other for a change in slope as well as intercept. We derive the asymptotic distribution of the tests as well as that of the feasible point optimal test (PF-Ls) which allows us to find the power envelope. The asymptotic critical values of the tests are tabulated and we compute the non-centrality parameter used for the local GLS detrending that permits the tests to have 50% asymptotic power at that value. Two methods to select the break point are analyzed. A first method estimates the break point that yields the minimal value of the statistic. In the second method, the break point is selected such that the absolute value of the t-statistic on the change in slope is maximized. We show that the MGLS and PTGLS tests have an asymptotic power function close to the power envelope. An extensive simulation study analyzes the size and power of the tests in finite samples under various methods to select the truncation lag for the autoregressive spectral density estimator. In an empirical application, we consider two U.S. macroeconomic annual series widely used in the unit root literature: real wages and common stock prices. Our results suggest a rejection of the unit root hypothesis. In other words, we find that these series can be considered as trend stationary with a broken trend. Given the fact that using the GLS detrending approach allows us to attain gains in the power of the unit root tests, a natural extension is to propose this approach to the context of tests based on residuals to identify cointegration. This is the objective of the second paper in the thesis. In fact, we propose residual based tests for cointegration using local GLS detrending to eliminate separately the deterministic components in the series. We consider two cases, one where only a constant is included and one where a constant and a time trend are included. The limiting distributions of various residuals based tests are derived for a general quasi-differencing parameter and critical values are tabulated for values of c = 0 irrespective of the nature of the deterministic components and also for other values as proposed in the unit root literature. Simulations show that GLS detrending yields tests with higher power. Furthermore, using c = -7.0 or c = -13.5 as the quasi-differencing parameter, based on the two cases analyzed, is preferable. The third paper is an extension of a recently proposed method to detect outliers which explicitly imposes the null hypothesis of a unit root. it works in an iterative fashion to select multiple outliers in a given series. We show, via simulation, that under the null hypothesis of no outliers, it has the right size in finite samples to detect a single outlier but when applied in an iterative fashion to select multiple outliers, it exhibits severe size distortions towards finding an excessive number of outliers. We show that this iterative method is incorrect and derive the appropriate limiting distribution of the test at each step of the search. Whether corrected or not, we also show that the outliers need to be very large for the method to have any decent power. We propose an alternative method based on first-differenced data that has considerably more power. The issues are illustrated using two US/Finland real exchange rate series.

本文的主要目的在於探討亞洲國家央行外匯存底的需求，研究的國家包括韓國、大陸、印度、新加坡、台灣與日本。透過使用1987年Engle-Granger的共整合分析法，我們發現這六國的外匯存底需求與其解釋變數具有共整合的現象。共整合現象的存在反映了這些國家的外匯存底需求存在長期的均衡關係。此外，為了進一步了解短期經濟的干擾如何影響外匯存底的需求，本文採用誤差修正模型做為分析工具，與過去文獻不同的是，本文的實證結果顯示本文所研究的六個國家之外匯存底顯著的受到貨幣成長率的影響，依據誤差修正模型的調整項亦可發現當外匯存底需求偏離長期均衡值時，此六國的調整速度相當慢，絕對值都在0.5以下，由於當一個經濟體系允許以緩慢的修正速度調整至長期均衡，必須擁有大量的外匯存底，由於本文所挑選的國家為全球著名的外匯存底持有國，其央行皆持有巨額的外匯存底，故本文的實證結果與理論假設一致。 / The primary purpose of this paper is to explore the demand characteristics for international reserves in some Asian countries including Korea, China, India, Singapore, Taiwan, and Japan. After applying the cointegration test of Engle-Granger (1987), we discover that the non-stationary macro time series of the group of the countries under study are cointegrated. Hence, international reserves in these Asian countries have displayed a long-run relationship with some determinants for the past several decades. Besides, we adopt an error correcting mechanism specification to investigate the short-run dynamic process of reserve holdings. Based on the error correction model (ECM), the rate of monetary growth is found to have a significant effect on reserve holdings in all of the six countries. In the end, the properties of the error correction terms among these countries are examined. We find that the absolute value of the error correction term is less than 0.5 in these economies. It implies that all of these six countries have very low adjustment speed of the demand for reserves. Owing to the huge stock of reserve holdings in this area, the results appear to be sensible.

<p>In recent years, two estimators have been proposed to correct the bias exhibited by the leastsquares (LS) estimator of the lagged dependent variable (LDV) coefficient in dynamic regression models when the sample is finite. They have been termed as &lsquo / mean-unbiased&rsquo / and &lsquo / medianunbiased&rsquo / estimators. Relative to other similar procedures in the literature, the two locationbased estimators have the advantage that they offer an exact and uniform methodology for LS estimation of the LDV coefficient in a first order autoregressive model with or without exogenous regressors i.e. ARX(1).</p> <p><br /> However, no attempt has been made to accurately establish and/or compare the statistical properties among these estimators, or relative to those of the LS estimator when the LDV coefficient is restricted to realistic values. Neither has there been an attempt to&nbsp / compare their performance in terms of their mean squared error (MSE) when various forms of the exogenous regressors are considered. Furthermore, only implicit confidence intervals have been given for the &lsquo / medianunbiased&rsquo / estimator. Explicit confidence bounds that are directly usable for inference are not available for either estimator. In this study a new estimator of the LDV coefficient is proposed / the &lsquo / most-probably-unbiased&rsquo / estimator. Its performance properties vis-a-vis the existing estimators are determined and compared when the parameter space of the LDV coefficient is restricted. In addition, the following new results are established: (1) an explicit computable form for the density of the LS estimator is derived for the first time and an efficient method for its numerical evaluation is proposed / (2) the exact bias, mean, median and mode of the distribution of the LS estimator are determined in three specifications of the ARX(1) model / (3) the exact variance and MSE of LS estimator is determined / (4) the standard error associated with the determination of same quantities when simulation rather than numerical integration method is used are established and the methods are compared in terms of computational time and effort / (5) an exact method of evaluating the density of the three estimators is described / (6) their exact bias, mean, variance and MSE are determined and analysed / and finally, (7) a method of obtaining the explicit exact confidence intervals from the distribution functions of the estimators is proposed.</p> <p><br /> The discussion and results show that the estimators are still biased in the usual sense: &lsquo / in expectation&rsquo / . However the bias is substantially reduced compared to that of the LS estimator. The findings are important in the specification of time-series regression models, point and interval estimation, decision theory, and simulation.</p>

Made available in DSpace on 2015-03-05T18:57:20Z (GMT). No. of bitstreams: 0 Previous issue date: 9 / Nenhuma / Essa dissertação procura avaliar a dinâmica da inflação no Brasil no período de março de 2000 a dezembro de 2007, através de uma Curva de Phillips híbrida, na especificação que permite além do termo “forward looking”, representado pela expectativa de inflação, o termo “backward looking” através da inflação defasada. Inicialmente, procura-se fazer uma exposição da teoria da Curva de Phillips, partindo da abordagem clássica até a abordagem dos novos keynesianos. A revisão da literatura destaca que a Curva de Phillips tem dificuldade em se ajustar à dinâmica da inflação. Dessa forma, a inclusão do termo que representa a inércia da inflação melhora sua adequação. Conjuntamente, o nível de atividade representado pelo hiato do produto não é estatisticamente significativo, fato esse que é suportado por outros trabalhos na literatura. Avalia-se a utilização do custo unitário do trabalho como proxy do nível de atividade, e os resultados não se alteram significativamente. Os dados de inflação relativos ao último trimes / This dissertation assess the inflation dynamics in Brazil through a hybrid Phillips Curve, at the specification that allows the "forward term", represented by inflation expectation, and the "backward term" through by inertial inflation. Initially, several approaches for the Phillips Curve's theory, from classical to new Keynesian, are presented and detailed. As pointed out by the precedent literature, it is shown that the Phillips Curve has some difficulty in adjusting to the inflation dynamics and, as a consequence, that the addition of the “backward term” enhances its fit. Furthermore, the inclusion of a proxy variable for the level of activity, represented by the output gap, is not statistically significant, result supported by previous studies in literature. Finally, it is analyzed the use of a unit labor cost as a proxy for the level of activity, with no significant changes in the results. The inflation data regarding the last tree months of 2002 are influenced by the effects of the political crisis at t

In time series models, autoregressive processes are one of the most popular stochastic processes, which are stationary under certain conditions. In this study we consider nonstationary autoregressive models of order one, which have iid random errors. One of the important nonstationary time series models is the unit root process in AR (1), which simply implies that a shock to the system has permanent effect through time. Therefore, testing unit root is a very important problem. However, under nonstationarity, any estimator of the autoregressive coefficient does not have a known exact distribution and the usual t &ndash / statistic is not accurate even if the sample size is very large. Hence,Wiener process is invoked to obtain the asymptotic distribution of the LSE under normality. The first four moments of under normality have been worked out for large n. In 1998, Tiku and Wong proposed the new test statistics and whose type I error and power values are calculated by using three &ndash / moment chi &ndash / square or four &ndash / moment F approximations. The test statistics are based on the modified maximum likelihood estimators and the least square estimators, respectively. They evaluated the type I errors and the power of these tests for a family of symmetric distributions (scaled Student&rsquo / s t). In this thesis, we have extended this work to skewed distributions, namely, gamma and generalized logistic.

<p>In recent years, two estimators have been proposed to correct the bias exhibited by the leastsquares (LS) estimator of the lagged dependent variable (LDV) coefficient in dynamic regression models when the sample is finite. They have been termed as &lsquo / mean-unbiased&rsquo / and &lsquo / medianunbiased&rsquo / estimators. Relative to other similar procedures in the literature, the two locationbased estimators have the advantage that they offer an exact and uniform methodology for LS estimation of the LDV coefficient in a first order autoregressive model with or without exogenous regressors i.e. ARX(1).</p> <p><br /> However, no attempt has been made to accurately establish and/or compare the statistical properties among these estimators, or relative to those of the LS estimator when the LDV coefficient is restricted to realistic values. Neither has there been an attempt to&nbsp / compare their performance in terms of their mean squared error (MSE) when various forms of the exogenous regressors are considered. Furthermore, only implicit confidence intervals have been given for the &lsquo / medianunbiased&rsquo / estimator. Explicit confidence bounds that are directly usable for inference are not available for either estimator. In this study a new estimator of the LDV coefficient is proposed / the &lsquo / most-probably-unbiased&rsquo / estimator. Its performance properties vis-a-vis the existing estimators are determined and compared when the parameter space of the LDV coefficient is restricted. In addition, the following new results are established: (1) an explicit computable form for the density of the LS estimator is derived for the first time and an efficient method for its numerical evaluation is proposed / (2) the exact bias, mean, median and mode of the distribution of the LS estimator are determined in three specifications of the ARX(1) model / (3) the exact variance and MSE of LS estimator is determined / (4) the standard error associated with the determination of same quantities when simulation rather than numerical integration method is used are established and the methods are compared in terms of computational time and effort / (5) an exact method of evaluating the density of the three estimators is described / (6) their exact bias, mean, variance and MSE are determined and analysed / and finally, (7) a method of obtaining the explicit exact confidence intervals from the distribution functions of the estimators is proposed.</p> <p><br /> The discussion and results show that the estimators are still biased in the usual sense: &lsquo / in expectation&rsquo / . However the bias is substantially reduced compared to that of the LS estimator. The findings are important in the specification of time-series regression models, point and interval estimation, decision theory, and simulation.</p>

The five essays in this dissertation are all concerned with how commodity price uncertainty affects the valuation of real and financial assets.  Focusing on the stochastic process approximating the price process of the commodity, a time-inhomogeneous mean reverting process is suggested and used in the valuation of a pulp mill.  Also an analytic approximation and a parameter estimation procedure to a stochastic volatility option-pricing model are developed.  Generally, the large valuation differences and hedging errors that occur for different assumptions about the price process indicate the importance of an appropriately specified price process.  The dissertation provides examples of this. The question of whether commodity prices are mean reverting or follow a random walk is also studied.  Using a large database with close to 300 different commodities, econometric tests favour a random walk.  There are very few exceptions.  However, when applied to an option pricing model, the time-inhomogeneous mean reverting process gives smaller hedging errors than the traditional Black-Scholes model based on a random walk.  The results are therefore inconclusive, although mean reversion seems more predominant than econometric tests reveal. / Diss. Stockholm : Handelshögskolan, 2003

Submitted by Vagner Enrico Castilho Alves (vagneralves@gmail.com) on 2014-08-26T17:29:47Z No. of bitstreams: 1 MPFE_Dissertacao_Vagner Enrico Castilho Alves_VF.pdf: 1506529 bytes, checksum: 5f1fd3e54f60edb8c87d8542948d2b57 (MD5) / Approved for entry into archive by JOANA MARTORINI (joana.martorini@fgv.br) on 2014-08-26T18:38:27Z (GMT) No. of bitstreams: 1 MPFE_Dissertacao_Vagner Enrico Castilho Alves_VF.pdf: 1506529 bytes, checksum: 5f1fd3e54f60edb8c87d8542948d2b57 (MD5) / Made available in DSpace on 2014-08-26T18:40:31Z (GMT). No. of bitstreams: 1 MPFE_Dissertacao_Vagner Enrico Castilho Alves_VF.pdf: 1506529 bytes, checksum: 5f1fd3e54f60edb8c87d8542948d2b57 (MD5) Previous issue date: 2014-08-08 / Este trabalho procurar analisar a validade da Paridade do Poder de Compra entre regiões metropolitanas do Brasil através do Índice de Preços do Consumidor Amplo (IPCA). Para isso foram realizados testes de raiz unitária para modelos lineares e não lineares, sobre cinco grupos do IPCA: Índice Geral, Administrados, Bens Comercializáveis, Bens Não Comercializáveis e Alimentos no Domicílio. O banco de dados utilizado compreende o período de 1991 a 2013 e os testes foram realizados sobre 550 séries, comparando-se todos os pares possíveis de regiões. Sob o modelo linear, não foi possível validar a PPC para a maioria das séries através do teste de raiz unitária DF-GLS, o que é diferente do esperado, uma vez que a análise intranacional elimina os efeitos da taxa de câmbio e reduz a influência dos custos de transações sobre as condições de arbitragem. Já o resultado do modelo não linear, realizado através do teste de Kapetanios, confirmou a estacionariedade de 203 séries, de tal forma que podemos afirmar que a PPC é válida para praticamente todos os pares possíveis de regiões metropolitanas abrangidas pelo IPCA nos cinco grupos estudados. Além disso, é possível observar que as séries apresentam maiores desvios entre os anos de 1991 e 1994, período marcado por grande instabilidade macroeconômica no Brasil e de sucessivos planos econômicos que não funcionaram. Após o início do plano real, em 1994, a relação da variação de preços entre regiões apresenta menor volatilidade e uma convergência mais rápida. / This paper analyses the validity of Purchasing Power Parity (PPP) between metropolitan regions of Brazil through the Consumer Price Index (IPCA). For this, we conducted unit root tests for linear and nonlinear models, on five groups of the IPCA: General Index, Regulated Prices, Tradable Goods, Non Tradable Goods and Food at Home. The database covers the period of 1991-2013 and the tests were conducted on 550 series, comparing all possible pairs of regions. On the linear model, it was not possible to validate the PPP for most of the series through the DF-GLS unit root test, which was not expected since the intra-national analysis should eliminate the effects of exchange rate and reduce the influence of transaction costs on arbitrage conditions. However, the result of the linear model, done through the Kapetanios test, confirmed the stationarity of 203 series, such that is possible to validate the PPP for almost all pairs of metropolitan areas covered by the IPCA in the five studied groups. Moreover, one can observe that the series have large deviations between the years of 1991 and 1994, a period marked by great macroeconomic instability in Brazil and successive economic plans that have not worked. After the beginning of the Real Plan in 1994, the ratio of the change in prices between regions stabilizes, presenting a low volatility and a short term convergence.