makeparam.norm {norm} | R Documentation |
Does the opposite of getparam.norm
.
Converts a list of user-specified parameters to a parameter vector
suitable for input to functions such as da.norm
and em.norm
.
makeparam.norm(s, thetalist)
s |
summary list of an incomplete normal data matrix created
by the function prelim.norm .
|
thetalist |
list of normal parameters of the same form as one produced by
getparam.norm . If the list has two components, the first must be the
vector of means and the second must be the covariance matrix,
where means and covariances are expressed on the scale of the original
data. If the list has three components, the first must be the vector
of means, the second must be the vector of standard deviations, and
the third must be the correlation matrix.
|
normal parameter in packed storage, suitable for use as a starting
value for em.norm
, mda.norm
, or mdamet.norm
.
prelim.norm
and getparam.norm
.
data(mdata) s_prelim.norm(mdata) #do preliminary manipulations thetahat_em.norm(s) #compute mle thetahat_getparam.norm(s,thetahat,corr=T) #extract parameters thetahat$r #look at mle correlations thetahat$r[1,2]_.5 #tweak a parameter thetahat_makeparam.norm(s,thetahat) #convert to packed storage thetahat_em.norm(s,thetahat) #run EM again from new starting value