[Martinson Twumasi]

This is very interesting. Please after calculating the turning point, what is the stata command to generate the Kuznets U-shape. Thanks.

[Jorrit Gosens]

You mean juts a graph?

clear 
set obs 1000
gen x = _n
gen y = runiform()*x-runiform()*x*x

gen lnx =ln(x)
gen lnxlnx = ln(x)*ln(x)
reg y lnx lnxlnx

predict yhat, xb
twoway line yhat x

See https://www.stata.com/manuals13/p_predict.pdf
and https://www.stata.com/support/faqs/s...ct-and-adjust/

[Justine Borja]

Hello Statalisters!

I am seeking help on how to interpret a turning point of a U-shaped curve. The turning point is 86 percent and my variables are co2 emissions and openness.

Thanks,
Justine

[Carlo Lazzaro]

Justine:
1) the outcome of the community-contributed module -utest- tells you that the relationship between -openness- (your regressand) and -co2 emissions- (one of your predictor) is U-shaped (ie, the null is not rejected);
2) the turning point is located at 86% (which is a minimum) value of -co2 emissions- (and falls within the range of the predictor);
3) the relationship between -openness- (your regressand) and -co2 emissions- (one of your predictor) increases to the right of 86% (ie, from >86% onwards), as the slope of the first derivative becomes positive.

[Daniel Loaiza]

estimates, how can I calculate the inflection point when GDP has first difference applied? I mean, I have a significant coefficient value of 0.65 for GDP and -6.4 for GDP2, if I calculate -b/2a it gives me approximately 2, then if I apply exponential it gives close to 8. However, the GDP per capita of my panel is It ranges between 1,000 and 18,000. In several studies I have seen that they calculate it but I don't know how they apply the inverse transformation of the first difference. I hope you can help me. greetings

[Mustafa Gm]

Dear Stata users,

I am working on a panel dataset on Environmental Kuznets Curve (EKC) Hypothesis. I regress carbon emission on GDP and population density by using "xtmg" module to estimate a quadratic EKC model with Augmented Mean Group (AMG) estimator. My question is about estimating turning point of EKC. I use the following commands suggested by Nick Cox in his earlier post:

xtmg CO2 GDP GDP^2 POPDENSITY, aug robust

di -_b[GDP]/(2 * _b[GDP^2])

local tp = -_b[GDP]/(2 * _b[GDP^2])

twoway qfit CO2 GDP, xli(`tp')

The commands works well and I can create a scatter plot for the whole panel. So, can I also create EK curves with a scatter plot for each cross-section unit in the panel? Thanks in advance for your help.