In recent years, research on the panelunitroottest methodology has focused on how to consider correlation between cross sections. Representative examples of initial work are the pooled t-test (LLC test) proposed by Levin et al. (2002), the averaged t-test (IPS test) developed by Im et al. (2003), and the combination test developed by Maddala and Wu (1999), all of which are referred to as first-generation panelunitroot tests. Under the strong assumption of cross-section independence, these methodologies proved more powerful than the unitroottest applied to univariate time-series data. The data, however, often do not satisfy the assumption of cross-section independence; cross-country (or regional) data used in testing, for example, the purchasing power parity hypothesis, income convergence hypothesis, and current account stability are indicative. In addition, prior research has clearly shown that application of the panelunitroottest, despite the existence of cross-section dependence in the data, results in serious size distortions (O’Connell, 1998; Strauss and Yigit, 2003; Pesaran, 2007). Many researchers have proposed “second-generation panelunitroot tests,” which consider correlation between cross sections, to overcome this problem. 1
This paper examines the stationarity of real GDP per capita for 27 OECD countries during the period 1950 to 2004. Using ADF unitroottest on single time series, it is found that real GDP per capita series of most OECD countries have unitroot. This outcome, however, might be due to the generally low power of this test. The aim of this paper is to reconsider this issue by exploiting the extra information provided by the combination of the time-series and cross-sectional data and the subsequent power advantages of panel data unitroot tests. We apply the test advocated by Im, Pesaran and Shin (1997). The results overwhelmingly indicate that real GDP per capita series among OECD countries are nonstationary.
Abstract. This paper extends the unitroottest of Christopoulos and Leòn-Ledesma (2010) to accommodate not only structural breaks and non-linear mean reversion, but also the contemporaneous cross-sectional dependence commonly found in panel dataset. The proposed test presents good finite sample properties and its applications on four major ASEAN countries’ real exchange rates show that the unitroot hypothesis could be rejected, supporting their long-run Purchasing Power Parity (PPP) against the Chinese Yuan.
This paper investigates the role of economic, instructional and political factors in attracting Foreign Direct Investment (FDI) inflows in Indochina (Cambodia, Laos, and Vietnam CLV) economies. Using panelunit- roottest and Random effects on panel data for 16 years from 1996 to 2012 to examine significant determinants of FDI in Indochina, the paper takes into account economic factors (inflation rate, trade openness, market size), institutional factors (corruption and rule of law), and political factors (political stability, government effectiveness, regulatory quality, voice and accountability) to explores the role of these determinants. The results show that market size, government effectiveness, rule of law and political stability are statistically significant and have positive influence on inward FDI.
There are several features that distinguish the Choi (2001) test from above- mentioned panelunitroot tests. First, this test is devised for finite N as well as for infinite N, where N denotes the number of groups. Second, it is assumed that each series has different types of nonstochastic and stochastic elements. Third, there is flexibility in the length of time series whereby each series can be appeared in different number of time series. Fourth, this test also deals problems with some groups have a unitroot and the others do not. Thus, the Choi (2001) test can be used under more general assumptions than the panelunitroottest of Im et al. (2003) and Levin and Lin (1993) 4 . Moreover, as mentioned by the
countries during the period 1980Q1-2011Q4. After checking for stationarity of the different variables of the model using Panelunitroottest, we applied the panel FMOLS, PDOLS and PMGE method to estimate the long-run money demand function for six GCC countries. From an aggregated analysis, we found that income elasticity is around 0.5 (FMOLS−α 1 = 0.537, DOLS α 1 = 0.496 and PMGE α 1 = 0.616) which is in line with the Baumol–Tobin model in which the income elasticity has to be β 1 = 0.5 (Baumol, 1952); Tobin (1956). The estimated coefficient of interest rates, which represents the semi-elasticity, is negative and significant at 1% level of significance (FMOLS−α 2 = −0.04, DOLS α 2 = 0.03, and PMGE α 2 = 0.04). This is also in line with standard monetary theory (Friedman, 1956) as holding physical assets produce costs. The empirical literature using aggregated time series data. Panel cointegration tests provided evidence in favor of a stable long-run money demand function. Moreover, similar results were found in the disaggregated analysis (individual countries). The Granger non- causality test due to Toda and Yamamoto (1995) procedure shows evidence of a bidirectional causal relationship between money demand and income. At an individual level, the unique common results between the countries are the evidence of a unidirectional causality running from M2 to income. Finally, the overall results show that exchange rate does not affect long-run money demand functions of the six GCC countries. The purpose of this study is to demonstrate the importance of money demand in conducting a sound monetary policy because the central banker in GCC countries would needs to make sure the elasticities are stable throughout time. This is one of the several requirements of a successful monetary union. The stability of the money demand function plays a central role for the importance of money for the monetary policy; especially because the GCC countries are moving toward a single currency managed by a single central bank. The goal of having union monetary policy strategies in GCC countries would support the price stability because many of those countries have faced an increase in inflation since 2002, which was accompanied with oil price boom. In fact, inflation decreases the purchasing power of consumers in the GCC countries. This research is important in this period because the GCC countries are trying to move toward creating of a Monetary Council and a single currency.
About FDI and urbanization research literature in this area, Chinese scholars, Rong-Lin (2010) use social statistical software SPSS and data of 13 cities in Jiangsu Province, analyze the correlation of FDI and urban development, it shows it is obvious between FDI and the level of urbanization in different regions, FDI stock on promoting the role of urban development is greater than the incremental role in promoting FDI on urban development. Xiu-Yu and Hong- Quan (2009) used the panelunitroottest, cointegration and error correction models in FDI relations with urbanization of Guangdong Province, the result shows reciprocal causation relationship between FDI and urbanization in long-term, but FDI is not the factor of urbanization in short term. Fan (2011) analyzed cross-sectional data of urbanization rate and FDI, per capita GDP, the proportion of tertiary industry and other variables of Jiangsu, Zhejiang and Shanghai, the result shows it is significant that foreign direct investment contribute to urbanization in Jiangsu, Zhejiang and Shanghai, but its significant level is less than the proportion of tertiary industry and the impact of the per capita GDP to urbanization. Research by Kai-Ming and Cun-Zhang (2010) showed that it has long-term equilibrium relationship between FDI and urbanization, FDI is the Granger cause to improve the level of urbanization. Lin Ji and Lai-Ke (2013) analyze the relationship among Chinese FDI (1978-2011), the urbanization rate and economic growth by VAR model. Studies have shown that the presence of long-term equilibrium relationship between FDI, urbanization and economic growth, FDI is Granger cause of economic growth, and can effectively promote economic growth, but economic growth is not the cause of FDI, the study also found that economic growth improve the level of urbanization, but the level of urbanization is not Granger cause of economic growth, urbanization has a negative effect on economic growth in the short term, but has a positive effect on economic growth in the long term. Based on the timing data of Jiangxi Province from 1984 to 2010, Ji-Zeng (2013) thought industrialization is the primary means of advancing urbanization; market size and market openness is the main reason for attracting FDI, promoting urbanization on FDI inflows have reverse effect; the original power of economic growth is urbanization and regional investment.
simulation. The specification of this model has some advantages over the panelunitroottest developed by Levin and Lin. First, because SUR estimation take account of the cross sectional correlation of error terms, it provides more information when compared to the single-equation ADF and Levin and Lin (1992,1993) tests. Second, Equation 5 assumes that the lag structure between the cross-sections constituting the panel are heterogeneous. Assuming the presence of unit-specific lag structures eliminates misspesification problem and makes each of error terms to be white-noise. Determining an identical lag structure for the cross-sections constituting the panel would cause biased test statistics. However, in SURADF method, one lag length is enough for eliminating the serial correlation problem in each cross-sections. In sum, the specification allows the different auto-regression coefficients among the units. In this method, by removing the limitation of (𝜌 1 − 1) = (𝜌 2 − 1) = ⋯ = (𝜌 𝑁 − 1), the null hypothesis that all the series have unitroot and the
Furthermore, a problem arises from the fact that the above studies assume that time series data are cross-sectionally independent among countries. This assumption is rather restrictive. Pesaran (2007) argues that panelunitroot tests can lead to spurious conclusions if they fail to take account of significant degrees of the cross-section dependence. He proposes a new panelunitroottest that allows for such dependence and shows that its presence can make a difference in the results obtained with conventional panelunitroot measures and his new one.
The efficient market hypothesis states that security prices fully reflect all available information and that the price fluctuations are unpredictable. Unpredictability of returns is satisfied if stock prices follow a random walk, that is, stock prices are characterized by a unitroot. These results show that the markets in this region seem to be weak form efficient in linear sense, however linear test are not taken into consideration of nonlinearities and this can be seen as model misspecification. By applying nonlinear test, first of all we see that the data generating process is nonlinear. With respect to this information, we obtain the true results about the market efficiencies of these region namely emerging markets of Europe. In this respect we make two important contributions to this literature. First, we have taken into account the possible nonlinearities in conditional mean of the series in testing efficiency of these markets which is a deviation from the vast literature. The second one, we have used Ucar and Omay (2010) nonlinear panelunitroottest which increase the power of nonlinear unitroottest (One way to obtain a more powerful test is to pool the estimates from a number of separate series and then test the pooled values). Furthermore, this is the first time a nonlinear panelunitroottest is used in the market efficiency literature.
The results of Levin et al. (2002), Im et al. (2003), Breitung (2002) and Fisher-type panelunitroot tests suggest, most of these tests are unable to reject the null hypothesis of unitroot in levels, which means that LFDI, LEX and LGDP are non-stationary in levels, but results of panelunitroot tests in the first difference indicate that all variables are stationary after the first difference. In other words, data series are integrated of order one I (1). The results of SSA countries by following the same test as applied in Levin et al. (2002), Im et al. (2003), Breitung (2002) mentioned in Table 1. On the basis of the panelunitroottest results, which imply that the data series are non-stationary in level, at the second step, we proceed to test for the existence of a long-run relationship between variables by using panel co-integration test. Granger (1981) showed that when some series are integrated in order one they become stationary after the first differencing, but a linear combination of them is already stationary without differencing, they are said to be cointegrated which implies the existence of cointegration in panel data (Table 2).
In this paper a variety of tests for the panelunitroot are employed. The first group consists of tests that do not allow for structural changes in series, constituted by the Levin, Lin and Chu (LLC) test (Levin et al., 2002), the Breitung (Breitung, 2000) test, the Im, Pesaran and Shin (IPS) test (Im et al., 2003), Fisher-type tests that employ ADF and PP tests, (Maddala and Wu, 1999, and Choi, 2001) and Hadri tests (Hadri, 2000). The LLC test is based on orthogonalized residuals and on the correction by the ratio of the long-run to the short-run variance of each variable. Although the test has become a widely accepted panelunitroottest, it has a homogeneity restriction, allowing for heterogeneity only in the constant term of the ADF regression. The Breitung test assumes that all panels have an autoregressive parameter and a unitroot process in common. The IPS test is a heterogeneous panelunitroottest based on individual ADF tests and was proposed by Im et al. (2003) as a solution to the homogeneity issue. It allows for heterogeneity in both the constant and slope terms of the ADF regression. Maddala and Wu (1999) and Choi (2001) proposed an alternative approach employing the Fisher test, which is based on combining the P-values from individual unitroottest statistics such as ADF and PP. One of the advantages of the Fisher test is that it does not require a balanced panel. Finally, the Hadri test is a heterogenous panelunitroottest that extends the KPSS (Kwiatkowski-Phillips-Schmidt-Shin) test – outlined in Kwiatkowski et al. (1992) – to a panel with individual and time effects, as well as deterministic trends. It takes as its null hypothesis the stationarity of the series.
Innovation is at the core of fourth industrial revolution which is already under way. Both Sustainable growth and development depend on technological innovation. Traditional economic models/theories are now undermined because of new technologies like AI, automation,3D printing, robotics etc. Lack of innovation creates major socio-economic problems such as inequality, unemployment, poverty and many more. Therefore, in this competitive world, a country needs innovative people with innovative ideas to go forward. The aim of this study is to explain and critically examine the determinants of technological innovation across 5 South Asian countries using yearly data for 1980-2015 period. This paper employs several econometric techniques such as Cross sectional dependence to see if shocks that occur in one country affect another, Panelunitroottest to check the stationary of the data and Panel Cointegration test to check long run relationship among the variables. This study also applies Fully Modified OLS to estimate long run coefficients and Dumitrescu and Hurlin panel causality test (2012) to see the causality between the variables. The findings suggest that democracy and human capital are negatively related to innovation, contrary to popular belief. The analysis also reveals that trade openness positively and significantly affects innovation and there exists a nonlinear, in particular an inverted U shaped relationship between innovation and financial development in South Asia. Findings from the Causality test reveals that there is bidirectional causality between total patent application and trade openness and also between financial development and human capital. This study, therefore, has several policy implications for South Asian countries.
ABSTRACT: This paper empirically examines short- and long-run relationships between foreign direct investments (FDI) and volatility of foreign portfolio investments (FPI) in 12 Central and Eastern European (CEE) countries. We use the Generalized Autoregressive Conditional Heteroskedasticity models to calculate volatility of the FPIs. We utilize the second generation panelunitroottest, panel- Wald causality test procedure and panel cointegration analysis allowing for structural breaks, and cross-sectional dependence. The results strongly suggest that a decrease in FPI volatility is followed by an increase in FDI in the long-run, and this indicates economies that advance in capital liberalization benefit from increases in FDI. However, the relationship in opposite direction in the long-run is valid in only half of the countries studied. In short-run, we observe that the former relationship is valid only in Turkey, the Czech Republic, and Lithuania; where the latter is valid only in Latvia.
be contemporaneously correlated (Pesaran, 2007). Cross sectional dependence is deemed to exist when there are strong co-movements among economic variables. As a result, studies that apply the first-generation panelunit-roottest may over-reject the stationary null and maybe overly supportive of PPP. In fact, O’Connell (1998) has emphasised that the true size of the test statistic can be far greater than the normal size when the underlying data generating process is characterised by cross-section dependence. This explains why evidence based on panel data set in the past tends to be more supportive of PPP. B-C in their work decided to tackle the issue by splitting the sample according to Asian and South and Latin American countries. For this reason, we formally test for the presence of cross-dependence (CD hereafter) and adopt a method that is specifically designed to handle this issue in panel setting. It should be noted that the literature has acknowledged that test statistic for the CD test, like the other diagnostic tests (e.g., nonlinearities), is also affected by extreme observations (and structural breaks). As such, we apply the CD test, not on the original RER series, but on the series after extreme observations during the currency episodes are removed from the data set before conducting the panelunit-root tests to confirm the PPP hypothesis. A large and growing body of research (e.g., Matsuki & Sugimoto, 2013; Baharum- shah et al., 2010; Nusair, 2008; Hooi & Smyth, 2007; Zurbruegg & Allsopp, 2004; Wu, Tsai, & Chen, 2004; Liew, Baharumshah, & Chong, 2004; Nusair, 2001) reveals that PPP holds in most but not all of the Asian countries when structural breaks are taken into account in the analysis. 4 All of them reported a major structural change that occurs
removed their regulatory measures at different stages of their economic development. Additionally, the deregulation process in these countries have varied in terms of timing and intensity (Phylaktis, 1999), with China being the last to enter the race following the country’s accession to the World Trade Organization (WTO). Being the sixth largest trading nation in the world and the second largest economy in Asia, China is widely believed to expand in international trading and continue to be the world’s fastest growing economies in the next decade 4 . Still, limited studies have actually looked at China’s connection with the other economies. Second, previous studies have relied on a number single-equation test to examine the unitroot null of RIP. Unlike these earlier works, we rely on recent advancements in the nonstationary panelunitroottest that allows for greater flexibility in modeling differences in the behavior across individual countries, and which has been proven quite satisfactorily in improving the power of the unitroot tests 5 . The low power of standard unitroot tests is one of the main motivations for the use of panelunitroot tests in recent work (see Im et al., 1997, on this issue) 6 . Nevertheless, unitroottest alone to examine if deviations from RIP (the RID series) are mean reverting is insufficient to testify the RIP condition. The speed of adjustment towards equilibrium parity rates is
develop a panelunitroottest that is robust to structural breaks due to currency crisis. They find that the long run PPP hold for the Asian countries, whereas the PPP relationship breaks down for countries of South and Latin America. While Coakley and Fuertes (1997) employ the IPS test only, Coakley, Kellard and Snaith (2005) apply IPS and CIPS (which takes the cross-sectional dependence into account) tests on two panels (CPI and PPI) of 15 OECD countries. Their results reject the PPP for the CPI panel, but for the PPI panel, PPP is not rejected. On the other hand, Drine and Rault (2007) apply panel cointegration technique to test for PPP. They form various panels, such as, OECD, African, Asian, Middle East and North Africa (MENA), Latin American, and Central and East European panels. They report favourable evidence for strong PPP in OECD panel and for weak PPP in MENA panel. For other panels, their study shows that PPP does not appear to characterise the long run behaviour of real exchange rates.
In order to analyze the stationarity of forward premiums and identify signi…cant historical events, we employ several types of unitroot tests that can detect structural breaks in data. A stationarity test was originally developed in order to check the time-series properties of univariate data (Dickey and Fuller 1979). Since then, much progress has been made in a number of directions, and Levin and Lin (1992) is one such example which proposed a panelunitroottest. Since researchers often face limited time-series observations, it is said that statistical power will be enhanced by incorporating cross-sectional information. Here the stationarity of forward premiums will be examined using the Lagrangian Multiplier (LM) based panelunitroottest (Im et al 2005) which is an extension of the LM unitroottest for univariate data (Lee and Strazicich 2003, 2004) and allows us to estimate endogenously the premium- speci…c timing of structural breaks.
Our empirical study is divided in three steps. The first step is to test whether the variables contain a panelunitroot to confirm the stationarity of M2, NOG (or GDP), Drate, Tbill, and Xrate. This is done by performing five type of panelunitroot tests which are: Levin-Lin-Chu (LLC, 2002), Im, Pesaran and Shin (IPS, 2003), the Augmented Dickey – Fuller (F-ADF), Philips – Perron (PP, 1998) and finally Breitung (2000). The second step is to check for panel cointegration tests using Kao (1999) and Pedroni (2004) to establish a cointegrating long-term equilibrium relationship between money demand and its determinants. Finally, the third step, we test for panel cointegration by using three different techniques: fully-modified ordinary least squares (FM-OLS) of Phillips and Hansen (1990) Kao and Chiang (2000), and Pedroni (2004) dynamic ordinary least squares (DOLS) estimator of Stock and Watson (1993) and canonical cointegrating regression (CCR) proposed by Park (1992).
of a unitroot could be rejected at that significance level during the reference period. Similar recursive exercises were carried out for the restricted euro area panel and for the changeover control panel. The lower part of Figure 2 shows the p-values of the recursive IPS tests. Results depend on which information criterion we use to determine the number of the lags in the test equations. Based on the AIC, which is the one for which results have been shown up to now, p- values for the restricted euro area panel are initially fairly stable and remained below the 5 percent significance level until inclusion of data prior to February 1990, i.e. after having added nearly six years of observations. When extending the period further in the past we can no longer reject the hypothesis of a unitroot, which suggests instability in the relation between actual inflation and quantified perceptions. However, once data for 1986 are included, p-values again approach the 10 percent significance level, and fell below it when data prior August 1986 are included. Results based on the Schwarz Information Criterion (SIC) (Schwarz, 1978) are much more in favor of stability for the restricted euro area panel, since the unitroot hypothesis can be rejected for all sample periods including pre-1996 observations. As to the changeover control panel, the unitroot is rejected for all sample periods including pre-1996 observations when the estimation is based on the SIC, whereas on the basis of the AIC some instability is found. P-values continuously oscillated around the 5 percent significance level until inclusion of data prior to 1988. Afterwards they clearly exceeded the 10 percent significance level. Note, however that in this experiment the control panel is formed by two countries only, as no Swedish data on perceptions are available prior to 1996. Moreover, for one of these countries (the UK) the null hypothesis of a unitroot was already not rejected during the reference period, when the AIC was used.