Dear Gretl Team, I'd tried to replicate the example 27.2 of the Gretl User’s Guide but I'd found an error:
<error message>
'Uhat' is not the name of a variable
Error executing script: halting
</error message>
I think the problem is only due to the name "muhat" and "Uhat".
Um abraço,
Henrique
################################
# #
# Example 27.2: Local level model #
# #
################################
function matrix local_level (series y)
/* starting values */
scalar s1 = 1
scalar s2 = 1
/* Kalman filter set-up */
kalman
obsy y
obsymat 1
statemat 1
statevar s2
obsvar s1
end kalman --diffuse
/* ML estimation */
mle ll = ERR ? NA : $kalman_llt
ERR = kfilter()
params s1 s2
end mle
return s1 ~ s2
end function
function series loclev_sm (series y, scalar s1, scalar s2)
/* return the smoothed estimate of \mu_t */
kalman
obsy y
obsymat 1
statemat 1
statevar s2
obsvar s1
end kalman --diffuse
series ret = ksmooth()
return ret
end function
/* -------------------- main script -------------------- */
nulldata 200
set seed 202020
setobs 1 1 --special
true_s1 = 0.25
true_s2 = 0.5
v = normal() * sqrt(true_s1)
w = normal() * sqrt(true_s2)
mu = 2 + cum(w)
y = mu + v
matrix Vars = local_level(y) # estimate the variances
muhat = loclev_sm(y, Vars[1], Vars[2]) # compute the smoothed state
foreign language=R --send-data
y <- gretldata[,"y"]
a <- StructTS(y, type="level")
a
StateFromR <- as.ts(tsSmooth(a))
gretl.export(StateFromR)
end foreign
append @dotdir/StateFromR.csv
ols Uhat 0 StateFromR --simple
Example27.2.inp
Description: Binary data
