The RATS Software Forum

option swhat in DLM is the same as output parameter state variance?

Like here swhat and vstates
dlm(c=1.0,sx0=1.e+7,sv=15099.0,sw=1469.1,presample=x1,y=nile,$
type=smooth,vhat=vhat,svhat=fhat,what=what,swhat=swhat) / $
xstates vstates

By test they are the same. Are there any traps here?

They shouldn't be the same, and, so far as I can tell, aren't. I get the following values from that example:

disp vstates(1)
4030.53277

disp swhat(1)
1468.88426