[Guest]

Dear River, yes that is exactly the paper I was referring to. I already have done the U test for the inverted U shape. If I am wrong it tells about the marginal effect - in my case bank credit to the private sector - on growth. Please correct me if I am wrong. One can read the output as:

Bank credit to the private sector is significant and positive at the minimum value of bank credit to the private sector and it is also negative but significant at the maximum value of bank credit to the private sector. Moreover, between 77% - 113% the relationship between credit to the private sector and growth is not statistically significant?

Thank you!

utest prcreditBI prcreditBI2 , fieller level(90)

Specification: f(x)=x^2
Extreme point: .9668481

Test:
H1: Inverse U shape
vs. H0: Monotone or U shape

-------------------------------------------------
| Lower bound Upper bound
-----------------+-------------------------------
Interval | .0704946 1.729018
Slope | 4.760985 -4.048266
t-value | 3.017159 -3.242765
P>|t| | .0017417 .000884
-------------------------------------------------

Overall test of presence of a Inverse U shape:
t-value = 3.02
P>|t| = .00174

90% Fieller interval for extreme point: [.77154139; 1.1332894]