Cross-sectional averages versus principal components

Westerlund, Joakim; Urbain, Jean‐Pierre

doi:10.1016/j.jeconom.2014.09.014

Cited by 119 publications

(104 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of PCs within the framework of this analysis has been documented by Westerlund & Urbain (2012, 2013b who have built on previous papers (Pesaran, 2006;Stock & Watson, 2002;Bai, 2003;Bai, 2009;Greenaway-McGrevy et al, 2012). As to what error are inherent in PC regressors, they have remarked on the possibility of normal inferences with PC-factors augmenting regressions, if the coefficients that are estimated converge toward their real values at the rate: NT , (where T is the number of time series and N, the number of cross sections).…”

Section: Principal Component Analysismentioning

confidence: 99%

Drivers of FDI in Fast Growing Developing Countries: Evidence from Bundling and Unbundling Governance

Asongu

Nwachukwu

2015

SSRN Journal

View full text Add to dashboard Cite

show abstract

Section: Principal Component Analysismentioning

confidence: 99%

Drivers of FDI in Fast Growing Developing Countries: Evidence from Bundling and Unbundling Governance

Asongu

Nwachukwu

2015

SSRN Journal

View full text Add to dashboard Cite

show abstract

“…Applying a similar logic to the calculation of idiosyncratic consumption, and approximating the common factors with cross‐sectional means of the variables, we obtain the more general model

c_{i t} - {normalγ}_{i}^{c} {true \overset{c}{true}}_{t} = α_{i} + β_{i} true(y_{i t} - {\tilde{normalγ}}_{i}^{y} {\bar{y}}_{t} true) + ϵ_{i t},

c_{i t} = α_{i} + β_{i} y_{i t} + {normalγ}_{i}^{c} {true \overset{c}{true}}_{t} + {normalγ}_{i}^{y} {true \overset{y}{true}}_{t} + ϵ_{i t},

where the β i coefficient measures the extent to which idiosyncratic shocks to income are channeled into idiosyncratic consumption. The approximation of common factors by cross‐sectional averages is advantageous for two reasons: first, the analysis of risk sharing focuses on consistent estimation of the β coefficient, but it does not concern itself with common factors per se, and second, Westerlund and Urbain () have shown that this approximation results in lower bias of the β estimate than competing approaches based on direct estimation of the common factors. The country‐specific

γ_{i}^{y} = - β_{i} {true \overset{γ}{true}}_{i}^{y}

and

{normalγ}_{i}^{c}

coefficients allow the amount of income and consumption driven by global shocks to vary across economies.…”

Section: An Alternative Approachmentioning

confidence: 99%

“…In the second stage, the CCE estimate of an individual β i is obtained by regressing the residual

{true \hat{ξ}}_{i t}^{c}

, capturing idiosyncratic consumption, on the residual

{true \hat{ξ}}_{i t}^{y}

, capturing idiosyncratic income. While the λ coefficients in (11) and (12) cannot be meaningfully interpreted (see Pesaran, ; Westerlund & Urbain, ), the residuals

{true \hat{ξ}}_{i t}^{c}

and

{true \hat{ξ}}_{i t}^{y}

are valid estimates of the idiosyncratic components and can be compared to cross‐sectionally demeaned consumption and income. Note that the latter may not be free of aggregate shocks: if the effect of global cycles differs across countries, cross‐sectional demeaning will not be able to isolate the idiosyncratic variation in the data and will therefore lead to biased conclusions about the extent of risk sharing.…”

Section: Empirical Strategymentioning

confidence: 99%

Common correlated effects and international risk sharing

Füleky

Ventura

Zhao

2017

International Finance

View full text Add to dashboard Cite

Correct assessment of consumption risk and its international diversification has important policy implications. However, existing studies of international risk sharing rely on the unrealistic assumptions that all economies are characterized by symmetric preferences and uniform transmission of global shocks. We relax these homogeneity constraints and compare our proposed approach with the conventional ones using a 44‐year panel of 120 countries. Our results confirm that consumption is only partially smoothed internationally and risk sharing is directly related to the level of development, but we do not detect a significant increase in risk sharing during the surge in financial globalization over the last four decades.

show abstract

“…The use of PC‐augmented estimators is also consistent with the above narrative. Westerlund and Urbain () have drawn from existing studies (Bai ; Greenaway‐McGrevy, Han, and Sul ; Pesaran ; Stock and Watson ) to elucidate factors derived from PCA. They have established that normal inferences are possible with PC regressors provided that the estimated coefficients converge at a rate

\sqrt{N T}

toward their real values.…”

Section: Methodsmentioning

confidence: 99%

Determinants of Growth in Fast‐Developing Countries: Evidence from Bundling and Unbundling Institutions

Asongu

2016

Politics & Policy

View full text Add to dashboard Cite

Purpose -We assess growth determinants in the BRICS (Brazil, Russia, India, China and South Africa) and MINT (Mexico, Indonesia, Nigeria and Turkey) fast-developing nations for the period [2001][2002][2003][2004][2005][2006][2007][2008][2009][2010][2011]. Particular emphasis is laid on the bundling and unbundling of ten governance dynamics.Design/methodology/approach-Contemporary and non-contemporary Fixed-and RandomEffects regressions are employed as empirical strategies. GDP growth and real GDP output are used as dependent variables. The governance variables are bundled by means of principal component analysis. Findings-The following are some findings. First, governance is more positively significant in non-contemporary specifications as opposed to contemporary regressions. Second, there is some interesting evidence on the heterogeneity of political governance as a driver. Political governance and its constituents (political stability and voice & accountability) are significantly positive in GDP growth but insignificant in real GDP output regressions. Third, the other governance dynamics are more significant determinants of real GDP output, as opposed to GDP growth. Accordingly, they are insignificant in contemporary regressions and negatively significant in non-contemporary regressions for GDP growth. Fourth, the constituents of economic governance have the highest magnitude in the positive effects of governance dynamics on real GDP output. Practical implications-The following are some practical implications. First, lag determinants are necessary for growth targeting or timing of growth dynamics. Growth drivers for the most part are more significantly determined by past information. Second, political governance is the most important driver of economic growth, with the significance of effects more apparent in non-contemporary regressions. Third, economic governance and institutional governance are more positively predisposed to driving real GDP output than GDP growth.Originality/value-As far as we have reviewed, it is the first study to investigate growth determinants in the BRICS and MINT nations. It has strong implications for other developing countries on the contemporary and non-contemporary dynamics of governance in driving economic prosperity.1 Simplice A. Asongu is Lead economist in the Research Department of the AGDI (asongus@afridev.org). 3JEL Classification: C52; F21; F23; O40 ; P37 Keywords: Economic Growth; emerging countries; governance AcknowledgementWe are highly indebted to Akpan Uduak for sharing his dataset.

show abstract

Cross-sectional averages versus principal components

Cited by 119 publications

References 6 publications

Drivers of FDI in Fast Growing Developing Countries: Evidence from Bundling and Unbundling Governance

Drivers of FDI in Fast Growing Developing Countries: Evidence from Bundling and Unbundling Governance

Common correlated effects and international risk sharing

Determinants of Growth in Fast‐Developing Countries: Evidence from Bundling and Unbundling Institutions

Contact Info

Product

Resources

About