The RATS Software Forum

It's sometimes a bad idea to copy code from a user's question. In:

equation mkteq mkt
# constant dy hhs(1,1) hhs(1,2) hhs(2,1) hhs(2,2)
equation iaeq ia
# constant dy hhs(1,1) hhs(1,2) hhs(2,1) hhs(2,2)

hhs(1,2) and hhs(2,1) are, by construction, identical so you can't use them both and have nothing to test. The correct set up for an "M" model that includes all the variances and covariances is:

equation mkteq mkt
# constant dy hhs(1,1) hhs(1,2) hhs(2,2)
equation iaeq ia
# constant dy hhs(1,1) hhs(1,2) hhs(2,2)

What would a11=a12 be testing?

You can export the variances/covariances quite simply as

open copy mygarchcovariances.xls
copy(format=xls,org=columns) / hhs

You can construct the correlations within RATS, but you can do it just as easily from the covariances in whatever the target application is for those.

TomDoan wrote:It's sometimes a bad idea to copy code from a user's question. In:

equation mkteq mkt
# constant dy hhs(1,1) hhs(1,2) hhs(2,1) hhs(2,2)
equation iaeq ia
# constant dy hhs(1,1) hhs(1,2) hhs(2,1) hhs(2,2)

hhs(1,2) and hhs(2,1) are, by construction, identical so you can't use them both and have nothing to test. The correct set up for an "M" model that includes all the variances and covariances is:

equation mkteq mkt
# constant dy hhs(1,1) hhs(1,2) hhs(2,2)
equation iaeq ia
# constant dy hhs(1,1) hhs(1,2) hhs(2,2)

What would a11=a12 be testing?

You can export the variances/covariances quite simply as

open copy mygarchcovariances.xls
copy(format=xls,org=columns) / hhs

You can construct the correlations within RATS, but you can do it just as easily from the covariances in whatever the target application is for those.

thanks again,
first, sorry for that
oh, question about H0, h11,h12,a11,a12 just example, to easier understand, H0: #3 = #4 =0, #5 = #6

I know I annoy you so much, sorry for that, but can you help me the last question,please..
how can I write code for: H0: #3 = #4 =0, #5 = #6

That's three restrictions (3=4=0 is "human shorthand" for 3=0 and 4=0).

restrict 3
# 3
# 1.0 0.0
# 4
# 1.0 0.0
# 5 6
# 1.0 -1.0 0.0