Dissertations / Theses on the topic 'Panel data unit root test'

There has been much interest in testing nonstationarity of panel data in the econometric literature. In the last decade, several tests based on the ordinary least squares and Lagrange multiplier methodhave been developed. In contrast to a unit root test in the univariate case,test statistics in panel data have Gaussian limiting distributions.This dissertation considers weighted symmetric estimation and maximum likelihood estimation in the autoregressive model with individual effects.The asymptotic distributions have been derived as the number of individuals and time periods become large. The power study from Monte Carloexperiments shows that the proposed test statistics perform substantiallybetter than those in previous studies even for small samples.As an example, we consider the real Gross Domestic Product per Capita for 12 countries.

In this thesis, I first investigate the relation between the aggregate unemployment rate and foreign direct investment (FDI) inflows and outflows. To study this relationship, I use a panel data set that contains 45 (developed and developing) countries observed from 1987 through 2008, and I employ Arellano and Bonds generalized methods of moments (ABGMM) estimation method for dynamic panel data. My results show that FDI inflows and outflows are not determinants of the aggregate unemployment rate. In addition, in line with macroeconomic theory, the previous level of aggregate unemployment has a positive impact on the current level of aggregate unemployment. Again, as macroeconomic theory suggests, my results show that per capita real gross domestic product (RGDP) has a negative effect on the current level of aggregate unemployment. Second, I study the long-run relationship between exports and per capita gross domestic product (instrumented by total population) using a panel data set of 51 countries from 1970 through 2008. To study this relationship, I employ the dynamic ordinary least squares (DOLS) estimation method. I find that the percentage of exports in nominal gross domestic products (GDP) is sensitive to changes in the populations of host countries and, hence, to the changes in their GDP. In addition, my results show that the agreement on trade related investment measures increased the percentage of exports in the nominal GDP of developed host countries more than it did in developing host countries.

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.

This thesis discusses some issues on unit root testing in panel data. It first examines the intra-China price convergence by employing panel unit root tests that take cross-sectional dependence into account. Contrast to the existing literature, where tests assuming independence are employed and PPP is found in the vast majority of goods/services prices, our study finds mixed evidence in favor of PPP. Mixed panels with both I(1) and I(0) units are then considered, a large scale simulation study is undertaken. Size/power of panel unit root tests are examined under a variety of DGPs. A battery of procedures designed for mixed panels are employed, and their performance are examined by simulation. An application on intra-China PPP shows that, on average, only a small proportion of stationary units can be found in relative price panels. We then consider fractionally integrated processes and propose two different types of panel fractional integration test, a Fisher-type test and a multiple testing procedure that controls the false discovery rate (FDR) and classify units into null and alternative. Simulation evidence is provided. Empirical application shows that, in our intra-China PPP study, strong evidence can be found against the unit root null.

Les pays africains de la zone CFA ont connu ces dernières années de multiples transformations économiques d'une part à travers les mesures initiées par les bailleurs de fonds bilatéraux et multilatéraux et d'autre part à travers les politiques d'intégration économique et monétaire. Ainsi, en partant de l'hypothèse selon laquelle du fait de ces nombreuses interventions, ces systèmes économiques incorporent divers phénomènes tels que les changements structurels et les dépendances inter-économies, nous avons étudié leurs principales implications sur la croissance, la convergence et la prévisibilité du taux de croissance. L'accent est d'abord mis sur les traits majeurs des politiques d'intégration dans le cadre d'une union monétaire tout en soulignant les éventuelles incidences de telles politiques sur la dynamique économique des pays membres principalement en termes de modélisation économétrique de la croissance et de la convergence. Les différentes études réalisées sur la base d'outils économétriques adaptés ont permis d'aboutir à des résultats nouveaux relatifs au processus de croissance et de convergence de ces économies comparativement à ceux basés sur les outils classiques de modélisation économétrique. Il ressort également de cette étude que la présence de facteurs communs et de ruptures structurelles est fortement liée aux politiques d'intégration mises en oeuvre au sein de la zone CFA. Ces résultats révèlent aussi que les chocs produisent des effets hétérogènes et ont généralement des dates d'occurrence différentes selon les pays et qu'il est nécessaire de faire varier les réponses de politique économique d'un pays à l'autre pour une croissance durable et mieux partagée.

This thesis is comprised of five papers that all relate to bootstrap methodology in analysis of non-stationary time series. The first paper starts with the fact that the Dickey-Fuller unit root test using asymptotic critical value has bad small sample performance. The small sample correction proposed by Johansen (2004) and bootstrap are two effective methods to improve the performance of the test. In this paper we compare these two methods as well as analyse the effect of bias-adjusting through a simulation study. We consider AR(1) and AR(2) models and both size and power properties are investigated. The second paper studies the asymptotic refinement of the bootstrap cointegration rank test. We expand the test statistic of a simplified VECM model and a Monte Carlo simulation was carried out to verify that the bootstrap test gives asymptotic refinement. The third paper focuses on the number of bootstrap replicates in bootstrap Dickey-Fuller unit root test. Through a simulation study, we find that a small number of bootstrap replicates are sufficient for a precise size, but, with too small number of replicates, we will lose power when the null hypothesis is not true. The fourth and last paper of the thesis concerns unit root test in panel setting focusing on the test proposed by Palm, Smeekes and Urbain (2011). In the fourth paper, we study the robustness of the PSU test with comparison with two representative tests from the second generation panel unit root tests. In the last paper, we generalise the PSU test to the model with deterministic terms. Two different methods are proposed to deal with the deterministic terms, and the asymptotic validity of the bootstrap procedure is theoretically checked. The small sample properties are studied by simulations and the paper is concluded by an empirical example.

Ogiltigt ISBN: 978-91-554-9069-0

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

Popularity of nonlinear threshold models and unit root tests has increased after the recent empirical studies concerning the effects of business cycles on macroeconomic data. These studies have shown that an economic variable may react differently in response to downturns and recoveries in a business cycle. Inspiring from empirical results, this thesis investigates dynamics of Turkish key macroeconomic data, namely capacity utilization rate, growth of import and export volume indices, growth of gross domestic product, interest rate for cash loans in Turkish Liras and growth of industrial production index. Estimation results imply that capacity utilization rate and growth of industrial production index show M-TAR type nonlinear stationary behavior according to the unit root test proposed by Enders and Granger (1998).

nÃo hÃ
The thesis consists of three chapters that have in common estimation models for panel data. The first chapter titled "Energy Consumption, GDP per capita and Exports: Evidence of long-term causality in a panel for the Brazilian States" analyzes the order of causality between the variables and then checks the long-term elasticities using the methodology FMOLS. It shows that GDP per capita is caused by their own past achievements, by consumption of electricity and exports. The consumption of electricity and exports, only are not caused by GDP per capita. Through the model FMOLS were estimated elasticities of long-term. The 1% increase in energy consumption and exports increased respectively 0.07% and 0.04 % in GPD per capita. The second chapter, entitled "Sustainability of Health Expenditure and Sanitation in Brazil: an analysis with Panel Data for the period 1985 to 2005" examines the sustainability of Health Expenditure and Sanitation of the states and the Federal District of Brazil, during the period 1985 to 2005. For this, we use the ratio of Expenditure by Function (Health and Sanitation) and GDP. The unit root tests for panel data refute the null hypothesis of presence of the unit root (the stochastic process is stationary) at 5% significance level. Accordingly, we can infer that the policy of health expenditure as a proportion of GDP remained almost stable (sustainable) over the period in question. The third chapter entitled "Formation of Convergence Clubs and Analysis of the Determinants of Economic Growth" support the formation of 10 clubs of convergence for a sample of 112 countries with per capita GPD data from 1980 to 2014 using the Phillips and Sul methodology (2007). Logged clubs and estimated a panel to investigate the impact of macroeconomic variables in the dynamics of economic growth rate through the Arellano and Bond model (1991) showed that: i) Inflation impacts the growth rate negatively, with effect greater for clubs that converge to a higher level of per capita income ii) imports as a proportion of GDP have positive relationship with the growth rate of per capita income for the countries belonging to clubs intermediaries, and a negative effect for other clubs iii) Exports as a proportion of GDP have a positive effect for all clubs, but is more pronounced for clubs that converge to a lower level of income and iv) international reserves have a positive effect for clubs that converge to high levels of income and a negative effect on clubs that converge to low levels of income.
A tese à composta por trÃs capÃtulos que possuem em comum modelos de estimaÃÃo para dados em painel. O primeiro capÃtulo intitulado âConsumo de Energia ElÃtrica, PIB per capita e ExportaÃÃo: Uma evidÃncia de causalidade de longo prazo em um painel para os Estados brasileirosâ analisa a o ordem de causalidade entre as variÃveis e posteriormente verifica as elasticidades de longo prazo atravÃs da metodologia FMOLS. Evidencia-se que o PIB per capita à causado pelas suas prÃprias realizaÃÃes passadas, pelo consumo de energia elÃtrica e pelas exportaÃÃes. Jà o consumo de energia elÃtrica e as exportaÃÃes, apenas nÃo sÃo causados pelo PIB per capita. AtravÃs do modelo FMOLS, estimaram-se as elasticidades de longo prazo. O aumento de 1% no consumo de energia e exportaÃÃes aumenta respectivamente 0,07% e 0,04% no PIB per capita. O segundo capÃtulo, intitulado âSustentabilidade dos Gasto com SaÃde e Saneamento no Brasil: uma anÃlise com Dados em Painel para o perÃodo de 1985 a 2005â examina a sustentabilidade dos gastos com saÃde e saneamento dos Estados e do Distrito Federal brasileiro, durante o perÃodo de 1985 a 2005. Para isso, utiliza-se da razÃo entre a Despesa por FunÃÃo (SaÃde e Saneamento) e o PIB. Os testes de raiz unitÃria para dados em painel refutam a hipÃtese nula de presenÃa de raiz de raiz unitÃria (i.e., o processo estocÃstico à estacionÃrio) ao nÃvel de 5% de significÃncia. Nestes termos, pode-se inferir que a polÃtica de gastos com saÃde como proporÃÃo do PIB praticamente permaneceu estÃvel (i.e., sustentÃvel) ao longo do perÃodo em questÃo. O terceiro capÃtulo intitulado âFormaÃÃo de Clubes de ConvergÃncia e AnÃlise dos Determinantes do Crescimento EconÃmicoâ sustenta a formaÃÃo de 10 clubes de convergÃncia para uma amostra de 112 paÃses com dados do PIB per capita de 1980 a 2014 atravÃs da metodologia Phillips e Sul (2007). Identificados os clubes e estimado um painel para verificar o impacto de variÃveis macroeconÃmicas na dinÃmica da taxa de crescimento econÃmico atravÃs do modelo Arellano e Bond (1991), evidenciou-se: i) A inflaÃÃo impacta a taxa de crescimento de forma negativa, com efeito maior para clubes que convergem para um nÃvel de renda per capita mais elevado; ii) As importaÃÃes como proporÃÃo do PIB possuem relaÃÃo positiva com a taxa de crescimento da renda per capita para os paÃses pertencentes a clubes intermediÃrios, e efeito negativo para os clubes do extremo; iii) As exportaÃÃes como proporÃÃo do PIB possuem efeito positivo para todos os clubes, porÃm à mais acentuado para clubes que convergem para um nÃvel de renda mais baixo e; iv) As reservas internacionais possuem efeito positivo para clubes que convergem para elevados nÃveis de renda e efeito negativo para os clubes que convergem para baixos nÃveis de renda.

The “special treatments” of automobile industry in China, especially in the forms of local protectionism, have been criticized as evidences of domestic market fragmentation for long. Whether these “special treatments” have stunted the integration of a national auto market in China remains a question. This paper seeks to examine the degree of market integration in the automobile markets in China by using tests of cointegration between prices of spatial markets. Several econometric approaches for spatial price analysis, including the ADF unit root test, Maddala-Wu’s Fisher type panel unit root test and more restrictive Dufour-Torres panel unit root test are applied to monthly average retail prices for the main models sold across 36 cities from 1994-2006. Besides the above conventional linear methods, the author also applies the newly developed nonlinear unit root method proposed by Kapetanios et al. (2003). Test results indicate that the nonlinear test support convergence more often than the conventional linear unit root tests. Moreover, they also reveal that price convergence and hence market integration hold for majority of models and markets. The paper also investigates possible explanatory factors in price disparities of auto markets among cities. As the evidence shows, the geographic distance between markets, difference of per capital income, and the existence of local production play important roles in the absolute price differentials as well as the volatility of price differentials among cities.

碩士
逢甲大學
經濟學所
91
Studies using standard unit-root tests generally cannot reject the hypothesis of a unit-root in unemployment rates. In this study, we test the hysteresis hypothesis in unemployment for Taiwan''s 23 regional data sets using panel-based unit-root tests for the June 1993 to October 2002 period. The results from SURADF test provide evidence that based on Taiwan''s regional unemployment data support the hysteresis hypothesis.

博士
國立成功大學
企業管理學系碩博士班
96
This thesis proposes test procedures to detect the presence of nonstationarity in dynamic panels against nonlinear but globally stationary exponential smooth transition autoregressive dynamic panels. The panels are assumed to be cross sectionally independent but each cross section is allowed to have its own short run dynamics and deterministic components. Based on whether the autoregressive parameters and speed of transition are identical or not across individuals, our panel models are classifies as homogeneous or heterogeneous. In homogeneous panels, we pool all cross section data to form a unit root test statistics after some normalization applied to each cross section data to control for the difference in short run dynamics and deterministic components. Whilst in heterogeneous panels, we first apply time series unit tests proposed by Kapetanios, Shin and Snell (2003) or Kapetanios and Shin (2003) to each cross section and then the panel unit root test statistics are formed by some transformation of these individual unit root test statistics. Two forms of transformation are used: one is the simple average of individual test statistics proposed by Im, Pesaran and Shin (2003) and the other is Fisher combined p-values statistics proposed in Maddala and Wu (1999) and Choi (2001). By using sequential limit theory, i.e., first T (the time series dimension) →∞ and then N (the cross section dimension) →∞, our test statistics are shown to converge in probability to a standard normal variate under the null of a common unit root. The finite sample properties of proposed test statistics are examined through Monte Carlo simulations. Overall, the simulation results indicate that our proposed test statistics performs well in terms of size and power. Application of the proposed test statistics to OECD real exchange rate data provides some evidence in favor of PPP hypothesis.

碩士
淡江大學
產業經濟學系碩士班
98
Since China has rapid growth by the open-door economic reforms of the late 1970s, a number of studies have argued that this outstanding economic growth might cause the unbalanced development with and within regions. In this study, we re-investigate whether provincial-level incomes in China continue to diverge or converge by the newly developed panel LM unit root test, which allows for considering structural breaks endogenously. In order to make a comparison, we also implement conventional methodology without consideration of breaks. The results show that it provides different outcomes when we consider structural breaks endogenously corresponding to, as Perron (1989) suggested, the misleading results if there is no consideration of structural breaks. More importantly, the results suggest that the income convergence should exist among Chinese provinces and that the occurrence of break timing. Furthermore, we apply panel KPSS unit root test, which has an opposite null hypothesis to panel LM unit root test, to examine the main results for providing more powerful and robust test statistics, and the outcomes indicate the consistent results. The result of existence of convergence in China is quite different from the public views. However, this might indicates that the policy of West Development eliminates the difference among provinces.

碩士
國立暨南國際大學
國際企業學系
98
In an efficient forward exchange market, the forward rate must be the unbiased estimator of future spot rate. Numerous researchers apply cointegration test to investigate the long-run relationship between spot exchange rate and forward exchange rate. Similarly, several other studies employ unit root test to examine the stationarity of forward premium. But less attention has been paid to the presence of structural breaks in the time series. This paper employs the panel unit root test proposed by Carrion-i-Silvestre et al. (2005) which consider multiple structural breaks to reexamine forward exchange market efficiency. The result demonstrate that the null hypothesis of panel stationarity for forward premiums of 15 countries can not be rejected, which in turn supports the hypothesis of foreign exchange market efficiency.

碩士
國立中山大學
經濟學研究所
103
This paper applies a newly-developed BCIPS panel unit root test that take smoothing structural breaks and cross-section dependence into account, proposed by Lee et al. (2015). We examine whether properties of the unemployment rates in twenty OECD countries can be explained by the natural rate hypothesis or hysteresis hypothesis, it will be able to provide an important reference for government policy. In the empirical results, the insufficient power in small number of samples appeared different results in the traditional unit root tests (ADF, PP and KPSS test). In the first and second generation panel data unit root tests, the IPS and LLC tests indicate that the unemployment rates in the OECD countries can be the stationary process; the CIPS test indicates that the unemployment rates in OECD countries can be the unit root process. However, first and second generation panel data unit root tests that do not take structural break into consideration may cause wrong results. The BCIPS test indicates that the unemployment rates in OECD countries can be the unit root process. Overall, my study concludes that the unemployment rates in OECD countries can be best described as unit root process in line with the hysteresis hypothesis.

碩士
淡江大學
財務金融學系碩士班
94
This paper applies random walk hypothesis to investigate whether stock-price indexes of nine Asia emerging markets can be characterized as random walk (unit root) or mean reversion processes. A more powerful test which allows for the presence of multiple structural breaks in the underlying series developed by Carrion-i-Silvestre, del Barrio and López-Bazo (2005) is employed. Results provide strong support for the random walk hypothesis.

This dissertation examines the causal relationship between inward foreign direct investment (FDI) and economic growth (GDP) in SADC countries. The study investigates, within a panel data context, whether causation is short-term, long-term or both; and explores whether the causal relationship between the two variables differs according to income level. The study covered a panel of 15 SADC countries over the period 1980-2012. In order to assess whether the causal relationship between FDI inflows and economic growth is dependent on the level of income, the study divided the SADC countries into two groups, namely, the low-income and the middleincome countries. The study used the recently developed panel data analysis methods to examine this causal relationship. It adopted a three stage approach, which consists of panel unit root, panel cointegration and Granger causality to examine the dynamic causal relationship between the two variables. Panel unit root results show that both variables in the two SADC country groups were integrated of order one. Panel cointegration tests showed that the variables for low-income country group were not cointegrated, while the variables for the middle-income countries were cointegrated. Since the low-income country group panels were not cointegrated, Grangercausality tests were conducted within a VAR framework, while causality tests for the middleincome country group were conducted within an ECM framework. Panel Granger causality results for the low-income countries showed no evidence of causality in either direction. However, for the middle-income countries’ panel, there was evidence of a unidirectional causal flow from GDP to FDI in both the long- and short- run. The study concludes that the FDI-led growth hypothesis does not apply to SADC countries. The results imply that the recent high economic growth rates recorded in the SADC region, especially middle-income countries, have been attracting FDI. In other words, it is economic growth that drives FDI inflows into the SADC region, and not vice versa. These findings have profound policy implications for the SADC region at large and individual countries.
Economics
MCOM (Economics)