As for the econometric analysis, I need more information in order to be able to tell whether the analysis is appropriate. In particular, you haven’t defined your variables clearly. In section 3, (page 10) you state that you considered 11 variables and created a data set covering 2000 to 2009. However, I’m not clear about a number of things.
• First of all, did you undertake a cross-section analysis so that you have one observation for each of the 47 countries? This is what you should be doing.
• Second, what is your dependent variable? Is it the average annual growth rate over the 2000-2009 period? It should be. You also need to specify whether the dependent variable is the per capita GDP growth rate or the per capita GNP growth rate; this is currently unclear.
• Third, you specify your first independent variable as “initial GYP per capita in 2000.” I presume this is the per capita GDP level in 2000. Or is it the per capita GNI in 2000? The GDP in 2000? You should be using either the per capita GDP in 2000, the per capita GNI in 2000, or the per capita GNP level in 2000 and whichever you use, it should be measured in constant PPP$. (I think in the World Bank data PPP$ may be referred to as International $. Last time I checked, they were either using 2000 or 2005 prices.) I assume GYP is just a typo or the code used for the indicator (if it stands for growth in income per capita it is not the variable you want—you want the level of GDP per capita in constant PPP$ in 2000).
• Fourth, you specify that you have included investment as a share of GDP. This is an appropriate indicator. You do need to make sure you average it over the 2000 to 2009 period. Also, you need to indicate whether you have entered it as a percentage (eg. 10% would get the value of 10 in your data set) or a proportion (eg. 10% would get the value of 0.10 in your data set). However, when I go to the estimated equation, I don’t see this variable.
• Fifth, you include savings. Firstly, you certainly don’t want to use the value of savings, if anything, you would want to use savings as a percentage of GDP. However, savings as a percentage of GDP will be roughly equal to investment as a percentage of GDP. Hence if you are going to include investment as a percentage of GDP (and that is hypothesized to be most tightly linked to growth) you do not want to include Savings as a percentage of GDP. To do so would give you a multicollinearity problem. I notice in the regression you shown the results for, you use savings and exclude investment. I recommend that instead you use the average annual investment share over the 2000 to 2009 period (That is, average your investment as a share of GDP data over the 2000-2009 period).
• Sixth, you specify the population growth rate. This is an appropriate variable. However, it should be averaged over the 2000-2009 period and you should specify how it is entered (is it entered as a percentage, eg. 3, or a proportion, eg. 0.03).
• Seventh, you indicate you have used the adult (15+) literacy rate as an
indicator of the level of human capital. You need to specify how you entered it—percentage or proportion. In your regression, you should either use the initial value of this (eg. 2000 value) or you should use the average over the 2000-2009 period. You need to be clear about what you did. Alternatively, if you find there is very little variation in that indicator, you could use the secondary school enrollment rate.
• Eighth, you indicate you have used Government revenue. First off, if you are going to use this variable, you should be using Government revenue as a percentage of GDP averaged over the 2000-2009 period. But you haven’t explained why you are using it. Is it your indicator of the size of the government sector? The more conventional indicator of the size of the government sector is Government Expenditures as a share of GDP. Why did you select Government revenue instead? It’s okay, but you will need to defend it.
• Ninth, you specify you used the ratio of exports to GDP as your trade openness indicator. It should be averaged over the 2000-2009 period. Clarify that it is. In addition, clarify how it is entered—percentage or proportion.
Use the order calculator below and get started! Contact our live support team for any assistance or inquiry.
[order_calculator]