Re: [Gretl-users] how to reproduce Verbeek's Table 5.4 (one-step GMM)?
On Wed, 16 Nov 2011, Allin Cottrell wrote:
[I see that Jack Lucchetti has replied on this since I started
writing, but
I'll post it anyway; having two methods to look at might be useful.]
Here's one more variant for comparison: this is closer in
spirit to Qi Shi's original approach. Note the "orthog"
specification
orthog elist ; const
will not work right unless you're using current CVS.
<hansl>
open pricing.gdt
set force_hc on
scalar delta=0.5
scalar gamma=0.5
series dcg = delta*cons^(-gamma)
list elist = null
series e0 = dcg*(1+rf)-1
elist = e0
loop j=1..10
series e$j = dcg*(rf-r$j)
elist += e$j
endloop
matrix V0=I(11)
gmm
dcg = delta*cons^(-gamma)
e0 = dcg*(1+rf)-1
e1 = dcg*(rf-r1)
e2 = dcg*(rf-r2)
e3 = dcg*(rf-r3)
e4 = dcg*(rf-r4)
e5 = dcg*(rf-r5)
e6 = dcg*(rf-r6)
e7 = dcg*(rf-r7)
e8 = dcg*(rf-r8)
e9 = dcg*(rf-r9)
e10 = dcg*(rf-r10)
orthog elist ; const
weights V0
params delta gamma
end gmm
</hansl>
Allin Cottrell