Problem with dynamic factor analysis: too few quotes r(132)

[Gre Vitor]

Hello mates.


I'm trying to run a factor dynamic analysis , but I'm having a problem. I am following the step by step of paper "Dynamic Factor Analysis with STATA".

Paper's link: http://www.stata.com/meeting/2italian/Federici.pdf

However, the following problem arises:

forvalues i=1/3{
mat e`i'=J(3,1,0)
}

forvalues i=1/3{
forvalues j=1/3{
mat e`i'[`j',1]=eigenvectors[`j',`i']
}
}

local t=1
matrix ST=J(3,3,0)

forvalues i=1(558)3348 {
matrix num=J(1,1,0)
matrix C=A[`i'..(`i'+558-1),1...]
svmat C
matrix accum cov=C1-C3, deviations noconstant
matrix cov=cov/(r(N)-1)
matrix ST=ST+cov
drop C1-C3
forvalues k=1/3{
mat num = num+(e`k" * cov * e`k')
}
mat l_`t'=num/trace(cov)
mat list l_`t'
local t=`t'+1
}

too few quotes
r(132);


I believe the problem is at following command line:
mat num = num+(e`k" * cov * e`k')

Could anyone help me, please?

I'm using STATA/MP 13.0.

Best regards.
Gre Vitor

[Robert Picard]

My guess is that the right quote used to expand the local macro k (i.e. `k') followed by a transpose operator (same right quote character) got converted to a double quote in the PDF. Try

mat num = num+(e`k'' * cov * e`k')

[Gre Vitor]

Dear Robert Picard,

You are right! I really need to thank you, your help was essential. It was a small mistake I made, I really could not find.

God bless you!

Best regards.

[Charlie Jiang]

Dear Gre Vitor
I am running a dynamic factor analysis,too.Still trying to figure out the procedure.
I don't understand when Federici wrote:


mat zi=J(1,3,0)
forvalues i=1/13 {
mat z=J(1,3,0)
forvalues k=`i'(13)117 {
mat z=z+A[`k',1...]
}
mat z=z/7
mat zi=zi\z
}
mat zi=zi[2...,1...]
mat list zi

" mat z=z/7" why 7 ? means time period(which is 9) minus 2?

Can you help me?

Thanks

Charlie Jiang