May be, this would be helpful
There are TWO issues: 1) NUMERIC
correctness and 2) ECONOMETRIC
correctness
1)
(i)
ols INF 0 PBI
eval  mean($uhat)

# 4.6556457e-14

# using multiple precision floating-point arithmetic,
# with the help of the Gnu Multiple Precision (GMP) library

mpols INF 0 PBI
eval  mean($uhat)

# -1.4953213e-14

The order of 10^-14 is a good approximation for zero

(I) A lss tested soft : Stata, Eviews, R (cran), R (Microsoft),
python give the same results

2) The regression is incorrect in ECONOMETRIC
sense as a consequence of at minimum 2 reasons:
(i) outliers for hyperinflation period (circa 1990)
(ii) PBI is obviously non-stationary
In the last case, besides summing up to zero,
the residuals could behave as strangely as they prefer to do

Oleh



2 березня 2017, 01:17:09, від "HEBERT SUAREZ CAHUANA" <hsuarezc@unsa.edu.pe>:

> I sent my dataset, I run regresion INF on PBI
>

>   _______________________________________________
> Gretl-users mailing list
> Gretl-users@lists.wfu.edu
> http://lists.wfu.edu/mailman/listinfo/gretl-users


_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users