Since the introduction of CUSUM statistic by E.S. Page (1951), detection of change or a structural break in time series has gained significant interest as its applications span across various disciplines including economics, industrial applications, and environmental data sets. However, many of the early suggested statistics, such as CUSUM or MOSUM, lose their effectiveness when applied to time series data. Either the size or power of the test statistic gets distorted, especially for higher order autoregressive moving average processes. We use the test statistic from Gombay and Serban (2009) for detecting change in the mean of an autoregressive process and show how the application of sieve bootstrap to the time series data can improve the performance of our test to detect change. The effectiveness of the proposed method is illustrated by applying it to economic data sets.

A recent outbreak of bird flu or avian influenza (AI), an especially highly pathogenic strain (HPAI) of H5N1, started in Hong Kong in January 2003 and caused 159 human deaths in Asia, Africa and Europe through early 2007. In addition, this outbreak resulted in millions of slaughtered birds and banned international trade of poultry meat in the infected countries. Such events harmed the poultry, tourism, and other related industries in the infected countries and changed the world poultry trade flow. Even in some uninfected countries, related industries were negatively affected. This study investigates the impact of bird flu outbreaks as manifested in financial markets within the US and Japan. The first essay explores how the avian influenza (AI) outbreaks impacted the security values of poultry-related firms. Using partial equilibrium analysis, this study infers that within a country AI outbreaks drop stock prices of poultry meat producers and raise stock prices of poultry food producers. Simultaneously, we infer that AI outbreaks in other poultry exporting countries raise stock prices of poultry meat producers and drop stock prices of poultry food producers. The empirical findings support our model results. Recent developments in time series method, directed graphs and search methods of cointegration rank are applied in this study. The second essay examines whether avian influenza outbreaks cause structural breaks in a model of their prices. It employs the dynamic programming algorithm and the reduced regression method for a cointegrated vector autoregressive (VAR) model to compute the break dates for the data sample. This research then compares the long run relation, and the short run relation and contemporaneous relation. The model estimations in these three sub-periods find these three sub-samples are significantly different. The breaks were caused by the invasion of Iraq on March 2003 and the 20 Bovine Spongiform Encephalopathy (BSE) induced ban of Canadian live cattle imports to the US on 03 March 2005, not by avian influenza outbreaks in early 2004. The third essay explores the effects of the avian influenza announcement in Japan on the prices of agricultural commodity futures contracts traded in Japan. Both the VAR model with asymmetric generalized autoregressive conditional heteroskedastic (GARCH) terms and the event study methods were used to examine whether avian influenza outbreaks significantly affected these markets. Our findings point out that the avian influenza outbreak only impacted the egg futures contract. These three essays found that outbreaks of avian influenza have significant impact on poultry-related stock prices and futures markets. The examined impacts changed the movement of those financial equity prices in the short run, but not in the long run. Research showed investors and poultry-related producers still encounter huge financial risk and loss.

The recent financial and economic turmoil driven by housing market has led the economists to refocus on the issue about monetary policy and asset price, especially housing price. In this dissertation I investigate the various relationships between monetary policy and asset prices in U.S. economy through steady state Bayesian VAR (SS BVAR) and revised Taylor-typed interest rate rule (Forward-looking rule) based on Generalized Method of Moments (GMM) methodology. In chapter II, steady state Bayesian VAR (SS BVAR) methodology is introduced and multi step-ahead forecasts are executed. Upon usual squared error loss methodology the forecasting performances of SS BVAR are evaluated in comparison with standard BVAR and conventional VAR. Equal predictive ability tests following Giacomini and White (2006) verify that the SS BVAR is superior in forecasting power especially in long-horizons. In chapter III, identification issue involving housing sector is explored through two different ways: economic theory-based approach and algorithms of inductive causations. Despite the different approaches the housing sector’s specifications are somewhat similar. Impulse response analyses demonstrate that monetary shock to housing price is relatively smaller, less significant, and less lasting when compared to Choleski identification. Also historical decomposition and conditional forecast analyses indicate that the housing price shock itself is crucial in accounting the sharp increase and sudden drop of housing price since 2003. Upon the estimated evidences I conjecture that there are much uncertainty between monetary policy and housing price, recalling the consideration of institutional factors when trying to accounting housing sectors. In chapter IV, following Dupor and Conley (2004), I explore how Fed responds to stock price and inflation movements differently across high and low inflation sub-periods. Replicated linear estimation results of Dupor and Conley (2004)’s indicate that Fed raises its target interest rate responding to stock price gap with statistical significance. Linear estimation results, however, are not robust to small change of chosen breakpoint especially in inflation coefficient. So I construct nonlinear model as an alternative way to relax this problem and carry out test of structural change with the nonlinear framework. Consequently both nonlinearity and structural change matter in explanation of Fed’s behavior in this type of reaction function analysis. Given structural change, inflation coefficients movement shows that Fed has responded to expected inflation pressure nonlinearly across sub-period, while stock price gap coefficient shows explicit break around early ’90 in line with Dupor and Conley (2004)’s finding.

Since the introduction of CUSUM statistic by E.S. Page (1951), detection of change or a structural break in time series has gained significant interest as its applications span across various disciplines including economics, industrial applications, and environmental data sets. However, many of the early suggested statistics, such as CUSUM or MOSUM, lose their effectiveness when applied to time series data. Either the size or power of the test statistic gets distorted, especially for higher order autoregressive moving average processes. We use the test statistic from Gombay and Serban (2009) for detecting change in the mean of an autoregressive process and show how the application of sieve bootstrap to the time series data can improve the performance of our test to detect change. The effectiveness of the proposed method is illustrated by applying it to economic data sets.

This thesis examines some of the economic effects associated with the elimination of single desk marketing on Canadian feed barley exports. It focuses on the interactions between Canadian and US spot feed barley markets in transition after this policy change in Canada. A brief introduction about world and regional barley markets is provided. The role of the CWB single desk and its role in Canadian barley marketing are discussed to motivate analysis about the effects of its absence. This study postulates there should be no significant change in Canada–US regional feed barley markets, based on conclusions from previous studies. This postulate is broken down into three testable hypotheses under the framework of spatial price analysis. With respect to the Canada–US regional feed barley market as single desk marketing was eliminated in Canada; (1) There should not be a significant structural break in the feed barley prices; (2) There should not be a significant change in market integration; (3) There should not be a significant change in the direction of price transmission. To test these hypotheses, the study employs econometric tests on Canadian and US prices spot prices for substitutable feed barley. The hypotheses are tested using a structural break test, a cointegration test, a Granger causality test, and associated impulse response functions. Since structural break tests do not find significant breakpoints in the data, the first hypothesis cannot be rejected. Next, the sample is split into two subsamples at the date when single desk was eliminated. An Engle-Granger procedure and the Johansen procedure are used to test cointegrating relationships between the variables. The results do not allow us to reject the second hypothesis of no significant change in market integration. In contrast, the third hypothesis is rejected, as a significant change is uncovered using the Granger causality test. Simulated impulse responses are also consistent with this finding.

This thesis investigates the short term and the long term cointegration relations between European and Chinese, European and Russian stock markets, with a goal to define international diversification benefits. Whereas Russia and China are considered as developing countries, Europe represents a developed market. The period of study is from 1997 to 2018, which considers the global 2007-2008 financial crisis as a shift in the equilibrium.The static cointegration long run findings demonstrate scope for diversification benefits for the all observing markets over the whole period. With regard to the sub periods (before and after the global financial crisis), the outcomes suggest increase in cointegration relations between Europe and China after the crisis, indicating a more diversified portfolio for investors before the crisis. European and Russian financial time series show no changing in cointegration linkages after the crisis, suggesting scope for diversification gains before and after the crisis in the long run.The dynamic cointegration results, however, demonstrate episodic cointegrating relations over the whole period for the all markets. These findings also clear illustrate growth in cointegration linkages during the first year of the crisis for all samples, suggesting a less diversified portfolio during this period (for the short horizon investors), and supporting the financial contagion effect in the short run.Looking at static and dynamic results, we recommend combining both methods in order to make a clear conclusion about benefits from international diversification.

Depuis sa première adoption en 1990, la politique de ciblage de l'inflation est devenue la politique monétaire la plus adoptée par les banques centrales aussi bien dans les économies développées ou émergentes. Cette priorité stratégique des autorités monétaires nous a conduits à réfléchir sur l'efficacité de ce choix. L'objectif de cette thèse est alors étudier la performance économique de cette politique économique surtout en cas de ruptures structurelles. Le premier chapitre présente l'approche théorique de ce travail en mettant l'accent sur ce débat de performance et l'intérêt à adopter cette politique de ciblage d'inflation afin de maitriser sa volatilité. L'approche empirique de cette thèse cherche à vérifier la performance de ce passage. Le second chapitre traduit l'efficacité du nouveau régime en étudiant son effet sur la dynamique de l'inflation, la croissance et certains indicateurs agissant sur la conjoncture macroéconomique. Ce travail explore également l'existence des différents points de ruptures structurelles. Le troisième chapitre illustre l'existence de ces dates de changement correspondant au changement de régime. La présente étude montre d'un côté qu'il y a performance économique suite au changement de régime. D'un autre côté, les résultats prouvent l'existence de multitude points de changement de rupture structurelle qui coïncident avec la mise en œuvre de la nouvelle politique de ciblage d'inflation. / Since its first introduction in 1990, the policy of inflation targeting has become the most adopted monetary policy by central banks in both developed and emerging economies. This strategic priority of monetary policy has led us to think the effectiveness of this choice. The objective of this thesis is then to study the economic performance of this economic policy especially in the case of structural interruptions. The first chapter presents the theoretical approach of this work focusing on performance and the necessity of adopting this policy in order to control its volatility. The empirical approach of this thesis aims to verify the performance of this transition. The second chapter shows the effectiveness of the new system by studying its effect on the inflation dynamics, the growth and indicators which affect the macroeconomic environment. This work also explores the existence of different points of structural interruptions. The third chapter illustrates the existence of these date changes corresponding to the regime switching. The present study reveals on one side that there is economic performance following the regime change. On the other hand, the results confirm the existence of many structural break points which coincide with the implementation of the new policy of inflation targeting.

O objetivo do trabalho foi analisar os efeitos da exportação de soja pelo Porto de Santarém (PA) no processo de formação de preços no mercado interno avaliando a integração tradicional entre os mercados no porto de Paranaguá (PR) e no município de Sorriso (MT). O período analisado foi de 2004 a 2013, com valores diários, dividido em dois subperíodos tomando como base para a divisão o potencial ponto de quebra estrutural no nível das diferenças relativas (percentuais) de preços entre Sorriso (MT) e Paranaguá (PR). O primeiro subperíodo considerou os preços de janeiro/04 a março/07, caracterizado pela ausência ou pela adaptação do mercado às novas oportunidades surgidas com a entrada em operação efetiva do porto de Santarém (PA). Já o segundo foi de março/07 a dezembro/13, considerando volumes mais expressivos e consolidados de exportações pelo porto paraense. Os testes de cointegração mostraram que as séries de preços se mantiveram cointegradas nos dois períodos analisados, porém o resultado interessante foi que as estimativas dos coeficientes da regressão da relação de equilíbrio de longo prazo apresentaram elasticidades de transmissão de preços estatisticamente diferentes. Considerando a modelagem VEC, foi observada pela análise de decomposição da variância que houve relativa redução da influência de Paranaguá (PR) nas mudanças de preços em Sorriso (MT) no segundo subperíodo, porém, sua importância permaneceu elevada. Já através da análise de impulsos, constatou-se que as mudanças de preços em Paranaguá (PR) são transmitidas integralmente (elasticidade unitária) a Sorriso (MT) no primeiro período e em torno de 80% no segundo, evidenciando que os preços na região mato-grossense se tornaram menos voláteis. O trabalho concluiu que, dado o processo transitório em termos logísticos observados no Brasil, principalmente no que diz respeito à finalização das obras na BR-163 e expansão da capacidade de escoamento por Santarém (PA), os preços de Sorriso (MT) ainda se mantém integrados ao porto de Paranaguá (PR). Existem, porém, várias evidências que apontam mudanças na formação de preço de Sorriso (MT) após a intensificação das exportações por Santarém (PA). / The objective was to analyze the effects of soybean exports through the Port of Santarem (PA) in the price formation process at the domestic market. It was evaluating the integration prices of two traditional markets: the port of Paranagua (PR) and Sorriso (MT). The study period was from 2004 to 2013, with daily values, divided into two sub-periods according to the potential structural break point at the level of relative prices differences (percentage) between Sorriso (MT) and Paranagua (PR). The first sub-period considered prices from January/04 to March/07, characterized by the absence or adaptation to new market opportunities from the effective beginning port of Santarem (PA) operation. The second was from March/07 to December/13, considering most significant and consolidated volumes exports by port of Santarem. The cointegration tests showed that the price series remained cointegrated in both periods analyzed, but an interesting result was the estimates regression coefficients of long-run equilibrium relationship showed transmission price elasticity statistically different. Considering the VEC model, was observed by analysis of variance decomposition that there was a relative reduction of the influence of Paranagua (PR) in price changes in Sorriso (MT) in the second sub-period, however, this importance remained high. By Pulse analysis, it was found that price changes in Paranaguá (PR) are transmitted in full (unit elasticity) to Sorriso (MT) in the first period and around 80% in the second, showing that prices at Mato Grosso region became less volatile. The work concluded that, given the transitional process in terms of logistics observed in Brazil, especially with regard to the completion of works on the BR-163 and expansion of the flow capacity by Santarem (PA), Sorriso (MT) prices still maintains integrated into the port of Paranagua (PR) . However, there are several evidences those changes in the formation of Sorriso (MT) price after the intensification of exports by Santarem (PA).

This paper analyzes effect of gun buyback that took place in Great Britain in years 1996 and 1997 on crime rate and compares the results with theoretical arguments and previous empirical findings. It contains analysis of three independent time series: crime rate in England and Wales, Scotland and Northern Ireland. Models of the time series are built using Box-Jenkins methodology. The models are tested for presence of a structural break using visual analysis, Chow test and Quandt-Andrews test. These tests are used as an evaluation criterion of the effect of buyback on crime rate. The result of the analysis is that it is not possible to reject the null hypothesis that buybacks do not have effect on crime rate.

The dissertation consists of three chapters on econometric methods related to parameter instability, forecasting and long memory. The first chapter introduces a new frequentist-based approach to forecast time series in the presence of in and out-of-sample breaks in the parameters. We model the parameters as random level shift (RLS) processes and introduce two features to make the changes in parameters forecastable. The first models the probability of shifts according to some covariates. The second incorporates a built-in mean reversion mechanism to the time path of the parameters. Our model can be cast into a non-linear non-Gaussian state-space framework. We use particle filtering and Monte Carlo expectation maximization algorithms to construct the estimates. We compare the forecasting performance with several alternative methods for different series. In all cases, our method allows substantial gains in forecasting accuracy. The second chapter extends the RLS model of Lu and Perron (2010) for the volatility of asset prices. The extensions are in two directions: a) we specify a time-varying probability of shifts as a function of large negative lagged returns; b) we incorporate a mean reverting mechanism so that the sign and magnitude of the jump component change according to the deviations of past jumps from their long run mean. We estimate the model using daily data on four major stock market indices. Compared to competing models, the modified RLS model yields the smallest mean square forecast errors overall. The third chapter proposes a method of inference about the mean or slope of a time trend that is robust to the unknown order of fractional integration of the errors. Our tests have the standard asymptotic normal distribution irrespective of the value of the long-memory parameter. Our procedure is based on using quasi-differences of the data and regressors based on a consistent estimate of the long-memory parameter obtained from the residuals of a least-squares regression. We use the exact local-Whittle estimator proposed by Shimotsu (2010). Simulation results show that our procedure delivers tests with good finite sample size and power, including cases with strong short-term correlations.