Unit Roots In Heterogeneous Panels Research Articles

Fiscal Policy and Economic Stabilization Dynamics in Sub-Saharan Africa: A New Evidence from Panel VEC Model and Hodrick-Prescott Filter Cyclical Decomposition

This study investigates the dynamics of fiscal policy and economic stabilization using a panel dataset from Sub-Saharan Africa (SSA) sourced from the World Development Indicators (WDI) between 1985 and 2019. The study utilized a panel vector error correction model (VECM) rather than panel vector autoregression (PVAR) after ascertaining the presence of cointegrating equations (existence of long-run relationships). Also, both homogenous and heterogeneous panel unit root processes were utilized to test the panel data’s stationarity before estimating the panel VECM. Again, the cyclical decomposition (Hodrick-Prescott Filter) is adopted in examining the effect of fiscal policy on economic stabilization across countries in SSA. Essentially, the VECM estimation result reveals that both government expenditure (GEX) and total reserves (TRS) have significant long-run relationships but are not substantial enough to stimulate economic growth and stability in the short run. Simultaneously, the Hodrick-Prescott Filter’s (HP-filter) cyclical decomposition confirms the non-stability of economic growth and stabilization in SSA over time. In line with other numerous policy recommendations, the SSA countries’ governments should improve on the policies targeted toward controlling the financial (fiscal) leakages to ensure more robust economic stabilization in the region through increases in total revenue and reserves both in the short and long-run.JEL Classification Codes: E62, E63, F38, F41, H50.

Reprint of: Testing for unit roots in heterogeneous panels

This paper proposes unit root tests for dynamic heterogeneous panels based on the mean of individual unit root statistics. In particular it proposes a standardized t-bar test statistic based on the (augmented) Dickey–Fuller statistics averaged across the groups. Under a general setting this statistic is shown to converge in probability to a standard normal variate sequentially with T (the time series dimension) →∞, followed by N (the cross sectional dimension) →∞. A diagonal convergence result with T and N→∞ while N/T→k,k being a finite non-negative constant, is also conjectured. In the special case where errors in individual Dickey–Fuller (DF) regressions are serially uncorrelated a modified version of the standardized t-bar statistic is shown to be distributed as standard normal as N→∞ for a fixed T, so long as T ¿ 5 in the case of DF regressions with intercepts and T ¿ 6 in the case of DF regressions with intercepts and linear time trends. An exact fixed N and T test is also developed using the simple average of the DF statistics. Monte Carlo results show that if a large enough lag order is selected for the underlying ADF regressions, then the small sample performances of the t-bar test is reasonably satisfactory and generally better than the test proposed by Levin and Lin (1993).

Reflections on “Testing for unit roots in heterogeneous panels”

This article is our personal perspective on the IPS test and the subsequent developments of unit root and cointegration tests in dynamic panels with and without cross-section dependence. In this note, we discuss the main idea behind the test and the publication process that led to Im et al. (2003).

Reflections on “Testing for Unit Roots in Heterogeneous Panels”

Reflections on “Testing for Unit Roots in Heterogeneous Panels”

Is there really hysteresis in the OECD unemployment rates? New evidence using a Fourier panel unit root test

We investigate the hysteresis hypothesis by proposing a heterogeneous panel unit root test that allows for gradually changing trends and cross-sectional dependence (CSD) among panel members using a flexible Fourier form. Inconclusive results from previous studies are potentially due to using very restrictive specifications with homogenous break structures and/or exogenously determined abrupt breaks. We seek to address these limitations by employing general specifications that are more capable of characterising the true data generation process of unemployment and by allowing for spill-over effects using a bootstrapping procedure to accommodate CSD that must be considered in a globalized world. Extensive simulations suggest that the failure to take structural breaks and CSD into account can lead to misleading conclusions about whether the unemployment rate is stationarity. We apply our test procedure to unemployment data for 23 OECD countries and find conclusive evidence against the hysteresis hypothesis for all these countries.

Rural-Urban Population Growth, Economic Growth and Urban Agglomeration in Sub-Saharan Africa: What Does Williamson-Kuznets Hypothesis Say?

The study analyses the dynamic effect of rural-urban relationship, economic growth and urban agglomeration in sub-Saharan Africa with a view of testing the validity of the Williamson-Kuznets hypothesis. The study utilized panel data analysis with pooled ordinary least squares with secondary annual time series data ranging from 1970 to 2017 and sourced from the World Bank database. The study equally employed both homogeneous and heterogeneous panel unit root tests to verify the stationarity of the panel data variables before estimating the model. However, the estimation result revealed that both rural and urban population growth has a negative impact as well as a statistically significant result on urban agglomeration in sub-Saharan Africa. The result equally showed a negative impact but statistically insignificant relationship between urban agglomeration and foreign direct investment. Also, a statistically significant and positive relationship was recorded between economic growth and urban agglomeration, thereby validating the Williamson-Kuznets hypothesis in sub-Saharan Africa. Based on the findings, the study among other numerous policy recommendations calls for a critical review of policies in the economies of sub-Saharan Africa to ensure effective utilization of the foreign direct investment net inflows towards initiating more and robust development projects both in the cities and rural areas, as well as expand the provisions of the basic infrastructural facilities and development projects. This would aim to curtail any perceived and unwarranted influx to the urban areas by the rural dwellers, hence they do not contribute significantly to growth in urban agglomeration.

How Does Governance Modify the Relationship between Public Finance and Economic Growth: A Global Analysis

How Does Governance Modify the Relationship between Public Finance and Economic Growth: A Global Analysis

Convergence of Health Expenditure in OECD Countries: Evidence from a Nonlinear Asymmetric Heterogeneous Panel Unit Root Test

This paper examines the convergence in health expenditure across 22 OECD countries between 1980 and 2012 by implementing panel unit root tests. Contribution of application of the nonlinear asymmetric heterogeneous panel unit root test is twofold. Firstly, it relaxes the assumption of cross-sectional dependency in panel data. Secondly, it incorporates the asymmetric nonlinear mean reversion in a panel setting. Results show that while the conventional panel unit root test cannot reject the null hypothesis of a unit root in relative per capita health expenditures for the whole set of countries, both the symmetric and the asymmetric nonlinear panel unit root tests indicate the stationarity of the panel. Specifically, almost 23 percent of the countries are found to be converging by employing the nonlinear asymmetric panel unit root test. In addition, introducing asymmetric structure helps to uncover additional converging countries which cannot be detected using linear and nonlinear symmetric panel unit root tests.

Unit root tests for cross-sectionally dependent panels: The influence of observed factors

This paper considers a heterogeneous panel unit root model with cross-sectional dependence generated by a factor structure—the factor common to all units being an observed covariate. The model is shown to be Locally Asymptotically Mixed Normal (LAMN), with the random part of the limiting Fisher information due to information generated by the covariate. Because of the LAMN structure, no asymptotically uniformly most powerful test exists; we investigate the asymptotic power properties of the locally optimal test, the best point optimal test, and the conditionally optimal t-test. Although one might expect the best point optimal test to be superior, the performance of the computationally simpler t-test is comparable.

Unit Root Tests for Cross-Sectionally Dependent Panels: The Influence of Observed Factors

This paper considers a heterogeneous panel unit root model with cross-sectional dependence generated by a factor structure – the factor common to all units being an observed covariate. The model is shown to be Locally Asymptotically Mixed Normal (LAMN), with the random part of the limiting Fisher information due to information generated by the covariate. Because of the LAMN structure, no asymptotically uniformly most powerful test exists; we investigate the asymptotic power properties of the locally optimal test, the best point optimal test, and the conditionally optimal t-test. Although one might expect the best point optimal test to be superior, the performance of the computationally simpler t-test is comparable.

Reexamining the PPP hypothesis: A nonlinear asymmetric heterogeneous panel unit root test

In this study, we re-examine the PPP hypothesis in the light of the new developments in the unit root testing literature. The recent theoretical findings have pointed out that the real exchange rate series exhibit asymmetric nonlinear behavior. A unit root test applied to analyze the PPP hypothesis therefore, should also take into account this asymmetry inherent in the real exchange rate. Different unit root tests that consider the presence of these data features have been developed in the time series literature. However, a true attempt to test the PPP hypothesis should take a panel data approach. To this end, we propose a nonlinear heterogeneous panel unit root test where the alternative hypothesis allows for symmetric or asymmetric exponential smooth transition autoregressive nonlinearity and provide its finite sample properties. We apply our test to the real exchange rates of the 15 European Union countries against the US dollar. While the results of the linear and symmetric nonlinear heterogeneous panel unit root tests are against the PPP hypothesis, the asymmetric nonlinear heterogeneous panel test that we propose gives support for the PPP hypothesis as expected. Therefore, the conclusions drawn from the linear panel unit root tests or the nonlinear panel unit root tests that do not take asymmetry into account might be misleading.

Unemployment hysteresis in the Eurozone area: evidences from nonlinear heterogeneous panel unit root test

Our article tests for hysteresis of unemployment rate for 17 Eurozone countries over the period 2000:1 to 2013:1 through the use of new nonlinear panel unit root tests by adopting the sequential panel selection method (SPSM), proposed by Chortareas and Kapetanios (2009) and a nonlinear panel unit root test developed by Ucar and Omay (2009). The results show that the unemployment rates for 17 Eurozone countries are not stationary and conform to the hysteresis hypothesis for both panel unit root tests of Ucar and Omay (2009) and SPSM without Fourier. We reported the results of the Panel Kapetanios, Shin and Snell (KSS) test with a Fourier function and found that the unemployment rates in 11 countries are stationary and this result is in accord with natural rate hypothesis. On the other hand, unemployment rates in six countries, namely Netherlands, Slovakia, Slovenia, Italy, Portugal and Cyprus, show hysteresis effect.

Environmental regulations and the export performance of South Korean manufacturing industries: A dynamic panel approach

This paper investigates the relationships among environmental regulation, export performance, and factor intensity, using panel data from South Korea's manufacturing sector (1991–2009). A panel vector autoregression (VAR) model in first differences tests the relationships among the variables, while considering the results of heterogeneous panel unit root and cointegration tests. Generalized Method of Moments (GMM) estimations help determine dynamic series relationships, and panel-causality tests are run based on the results of GMM estimations. There is evidence of a positive short-run linear causal relation running from environmental regulation to export performance, suggesting that environmental-protection expenditure may constitute a comparative advantage. The short-run linear causal relation from export performance to investment in activities related to environmental protection is insignificant.

TESTING FOR THE EFFICIENT MARKET HYPOTHESIS IN STOCK PRICES: INTERNATIONAL EVIDENCE FROM NONLINEAR HETEROGENEOUS PANELS

Using international data, this paper explores whether the efficient market hypothesis for real stock prices is supported for different panels. The stationarity of a real stock price has important implications for modeling and forecasting financial activities. On a global scale, we implement the recently developed nonlinear heterogeneous panel unit root test, which allows us to account for possible nonlinearity and cross-section dependence and to identify how many and which countries of the panel contain a unit root. The primary conclusion is that the stationarity of real stock prices varies between regions and levels of economic development. Overall, our empirical results illustrate that real stock prices in these countries are a mixture of stationary (integrated of order zero) and nonstationary (integrated of order one) processes.

TEST OF THE CONVERGENCE HYPOTHESIS ALLOWING FOR CROSS‐SECTIONAL DEPENDENCE

AbstractA large empirical literature has investigated whether per capita output converges around a common trend across national and regional economies. The methods used in this literature assume no cross‐sectional dependence even though it is likely to be present and might be important in practice. Chang has devised a promising method of testing for unit roots in heterogeneous panels with cross‐sectional dependence. We apply her method to test whether convergence takes place across three samples of economies: 15 advanced industrial countries; a broader group of 57 countries; and the 48 contiguous US states. We find evidence of convergence for the 15 advanced industrial economies but no evidence of convergence across either the broad group of countries or the US states.

Why panel tests of purchasing power parity should allow for heterogeneous mean reversion

Recent studies of purchasing power parity (PPP) use panel tests that fail to take into account heterogeneity in the speed of mean reversion across real exchange rates. In contrast to several other severe restrictions of panel models and tests of PPP, the assumption of homogeneous mean reversion is still widely used and its consequences are virtually unexplored. This paper analyzes the properties of homogeneous and heterogeneous panel unit root testing methodologies. Using Monte Carlo simulation, we uncover important adverse properties of the panel approach that relies on homogeneous estimation and testing. More specifically, power functions are low and assume irregular shapes. Furthermore, homogeneous estimates of the mean reversion parameters exhibit potentially large biases. These properties can lead to misleading inferences on the validity of PPP. Our findings highlight the importance of allowing for heterogeneous estimation when testing for a unit root in panels of real exchange rates.

Why Panel Tests of Purchasing Power Parity Should Allow for Heterogeneous Mean Reversion

Recent studies of purchasing power parity (PPP) use panel tests that fail to take into account heterogeneity in the speed of mean reversion across real exchange rates. In contrast to several other severe restrictions of panel models and tests of PPP, the assumption of homogeneous mean reversion is still widely used and its consequences are virtually unexplored. This paper analyzes the properties of homogeneous and heterogeneous panel unit root testing methodologies. Using Monte Carlo simulation, we uncover important adverse properties of the panel approach that relies on homogeneous estimation and testing. More specifically, power functions are low and assume irregular shapes. Furthermore, homogeneous estimates of the mean reversion parameters exhibit potentially large biases. These properties can lead to misleading inferences on the validity of PPP. Our findings highlight the importance of allowing for heterogeneous estimation when testing for a unit root in panels of real exchange rates.

A new score test for unit roots in heterogeneous panels — Residual likelihood approach

Using the marginal likelihood of the residual, we propose new score type unit root tests for heterogeneous panels. Our tests are more powerful than the t-bar test of Im et al. (2003) for panels with many cross-sectional units and short time spans.

Examining the behaviour of visitor arrivals to Australia from 28 different countries

Testing the integrational properties of visitor arrivals has important implications for policy, for if visitor arrivals are integrated of order one (nonstationary) then it implies that shocks to visitor arrivals are permanent. However, if visitor arrivals are found to be integrated or order zero (stationary) then this implies that shocks to visitor arrivals are temporary. In this paper we examine whether visitor arrivals to Australia are stationary or nonstationary, using the recently developed univariate and panel Lagrange multiplier tests, and the Im, Pesaran and Shin [Im, K.S., Pesaran, M.H., Shin, Y., 1997. Testing for Unit Roots in Heterogeneous Panels. Manuscript, Department of Applied Economics, University of Cambridge; Im, K.S., Pesaran, M.H., Shin, Y., 2003. Testing for Unit Roots in Heterogeneous Panels. Journal of Econometrics, 115, 53–74] panel t-test. Our exercise involves Australia’s 28 tourist source markets. Our main findings are: (1) that visitor arrivals to Australia from 28 tourist source markets are stationary, implying that any shock will have only a temporary effect and (2) the second structural break, which mainly coincides with the September 11 terrorist attacks and the Asian financial crisis, has slowed down the growth rate in visitor arrivals to Australia from 22 out of 28 (79%) of the tourism source markets.

Testing for seasonal unit roots in heterogeneous panels in the presence of cross section dependence

This paper presents two alternative methods for modifying the HEGY-IPS test in the presence of cross-sectional dependency. In general, the bootstrap method (BHEGY-IPS) has greater power than the method suggested by Pesaran [Pesaran, M.H. (2007). A simple panel unit root test in the presence of cross section dependence. Journal of Applied Econometrics, forthcoming.] (CHEGY-IPS), although for large T and high degree of cross-sectional dependency the CHEGY-IPS test dominates the BHEGY-IPS test.