Hi,
I have a problem with stata (I use stata/SE 9.2) : I want to run a xtpmg
regression (as my data are panels with time series and I want to deal with
the problems of heterogeneity and dynamics) and stata says me that
"initial values are not feasable". I saw you posted an answer to that
question in 2003, but stata says me that from() is not an option that is
accepted with xtpmg. Can you help me to solve this problem ?
The do I use is :
clear
set mem 500m
set maxvar 32767
set matsize 5000
ssc install xtpmg
use "C:\Documents and Settings\Julia Cag�\Mes documents\M�moire
M2\Mod�le\Test empirique\Fichier reg Africa julia6.dta"
encode(country), gen(countries)
gen fdi_gdp_ratio= fdi/gdp
gen nfdi_gdp_ratio= nfdi/gdp
/* regression investissement sur �pargne aide fdi*/
reg gfcf_gdp_ratio gds_gdp_ratio fdi_gdp_ratio oda_gdp_ratio
matrix b0 = e(b), ln(e(rmse)^2),.1
matrix list b0
drop if gfcf_gdp_ratio>=.
drop if gds_gdp_ratio>=.
drop if fdi_gdp_ratio>=.
drop if oda_gdp_ratio>=.
tsset countries year
/* pmg regression investissement sur �pargne aide fdi*/
xtpmg d.gfcf_gdp_ratio d.gds_gdp_ratio d.fdi_gdp_ratio d.oda_gdp_ratio if
year>=1963, ufrom(bO,copy) lr(l.gfcf_gdp_ratio gds_gdp_ratio fdi_gdp_ratio
oda_gdp_ratio) ec(ec) replace pmg
Thank you very much,
Best,
Julia
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
