Displaying 1 result from an estimated 1 matches for "ssm2".
Did you mean:
sse2
2009 May 29
0
possible bug in "sspir" package?
..."y") based on the rows of
the input matrix y rather than based on the columns. Thus, if I change
those lines and rerun the code above, I do in fact get an estimated
state vector of length=20 (and the estimates seem reasonable). My
modified version of your code is attached to this msg (ssm2.r).
I also have an additional question that may or may not be related to
this. If I fit a random-walk model using the approach outlined above and
my ssm2 function (assuming it is correct), the point estimates of the
state vector are nearly identical to the "observed" data. However, if...