EViews.com

Posted: **Mon Oct 14, 2013 2:46 am**

Hi,

I’m writing my first program and I have some difficulties with a very simple thing.
I want to run a gmm regression for different countries: us uk eu15 eu17 jp ger au .

I set my gmm equation and I would like to obtain different gmm estimate , one for each country . I have a list of series ,in this case labs and infl, for all these countries. The program I wrote always run the same regession with the same instruments for only one country ( it iterates several times only the same gmm equation for au(Australia)). I wrote

Code: Select all

open ppp.wf1

[color=#FFFF00]smpl 1960:1 2013:3

   For %j  a  au sw  us uk sp it  eu15 eu17 cn  nd fn jp ger

genr  infl{%j}=400*log(cp{%j}/cp{%j}(-1))
genr  inflpd{%j}=400*log(ipd{%j}/ipd{%j}(-1))
next

for %k1  us uk eu15 eu17 jp ger au

for %countries us uk eu15 eu17 jp ger au

genr  infl{%k1}=400*log(cp{%k1}/cp{%k1}(-1))
genr labs{%k1}=(ulc{%k1}/ipd{%k1})*0.123

labs{%k1}.hpf(lambda=1600) labs{%k1}_hp
series labs{%k1}_c = labs{%k1} - labs{%k1}_hp

equation {%countries}.gmm(method=converge, instwgt=HAC,instinfo=SIC,deriv=aa) infl{%k1} c infl{%k1}(+1) infl{%k1}(-1)  labs{%k1}_c@ infl{%k1}(-2) infl{%k1}(-3)  infl{%k1}(-4)  infl{%k1}(-5)  labs{%k1}_c(-1)  labs{%k1}_c(-2)  labs{%k1}_c(-3)   labs{%k1}_c(-4)
{%countries}.weakinst
next 
next

[/color]

What is wrong in the program?

I have to do a Bai- Perron test also. I did the “add in” for my eviews 7.1 and the test works for each single series for each single country , but how to make the program run it?

Thanks so much in advance for any kind of help!!

Posted: **Mon Oct 14, 2013 10:27 am**

You don't need the %countries loop and the %k loop. They are the same loop. Remove the %countries loop, and then call your equation eq_{%k}

Posted: **Tue Oct 22, 2013 8:55 am**

I'm a little bit late but thanks so much Gareth.

I have an other question: how can I perform a bai-perron test for each ols equation in eviews 7.1?
If I write

Code: Select all

For %j  a  sd  us uk  it  eu15  cn  nd fn jp ger
equation eq0_{%j}.ls  inflpd{%j} c inflpd{%j}(-1)
eq0_{%j}.baiperron
next

I only get a bai-perron test table for the first country.
What is again wrong?
I would be very grateful for any feedback you might be able to give me
In any case, thanks again for your answer!

EViews.com

gmm estimate for different countries

gmm estimate for different countries

Re: gmm estimate for different countries

Re: gmm estimate for different countries