My study uses annual data for variables of growth rate of the real GDP, growth rate of the real GDP per capita, the primary school enrolment, the secondary school enrolment, the tertiary school enrolment institutional capital and the labour force participation in order to analyze the impacts of female education on the economic growth. before estimating my model i want to check the stationarity of all these variable because the majority of the economic series is nonstationarity according to many empirical studies and in order to avoid the spurious regression. For this reason, we will use the panel unit root tests in order to check stationarity or nonstationarity in economic series.
many thanks
what You think of my reasoning?
Dear Aicha, your trouble about nonstationarity is correct, but also the comment of Cengiz Aytun is correct.
If your datas are nonstationary, I suggest you to refer to:
1) Phillips PC and Moon HR (2001) - Nonstationary panel data analysis: an overview of some recent developments, Cowles Foundation paper n. 1009
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.414.7173&rep=rep1&type=pdf
2) Blackburne EF and Frank MW (2007) - Estimation of nonstationary heterogeneous panels, Stata Journal, vol. 7, n. 2
http://www.shsu.edu/eco_mwf/Blackburne_Frank.pdf
3) Choi I (2007) - Nonstationary panels, in Volume 1 of the Palgrave Handbook of Econometrics: Theoretical Econometrics
http://www.homes.eco.auckland.ac.nz/esba001/InChoiRev.pdf
4) Smith RP (2001) - Estimation and inference of nonstationary panel time series data, Department of Economics, Birkbeck College, London (prepared for the RC33 Conference, Cologne, october 2000)
http://carecon.org.uk/UWEMasters/Applied%20Econometrics/panel1.pdf
5)Baltagi BH and Kao C (2000) - Nonstationary panels, cointegration in panels and dynamic panels: a survey
http://tyigit.bilkent.edu.tr/gradmetr/survey.pdf
Thank you
I have consulted some these references.
I found interesting results if I use the data stationnaire. the data used are the annual data (growth rate of the real GDP, growth rate of the real GDP per capita, the primary school enrolment, the secondary school enrolment, the tertiary school enrolment and the labour force participation) over the period 1980-2012. On the other hand, the variables measured the institutional capital, these variables are available only over period 2000-2012. The countries included are Algeria, Morocco, Egypt and Tunisia. The stationarity is tested with two generations of the panel unit root tests. my problem is for the institutional capital where the period is short. for these variables I found also that the series are nonstationary.
thank you
very interesting arguments especially when you wrote ''Especially for institutional variables, transformation is too slow for countries''.
many thanks
- Badges
- Science topic
- Similar topics
- Correction