makeparam.norm: Convert normal parameters to packed storage

In norm: Analysis of Multivariate Normal Datasets with Missing Values

Convert normal parameters to packed storage

Description

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.

Usage

makeparam.norm(s, thetalist)

Arguments

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.

Value

normal parameter in packed storage, suitable for use as a starting value for em.norm, mda.norm, or mdamet.norm.

See Also

prelim.norm and getparam.norm.

Examples

data(mdata)
s <- prelim.norm(mdata)   #do preliminary manipulations
thetahat <- em.norm(s)   #compute mle
thetahat <- getparam.norm(s,thetahat,corr=TRUE)   #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

norm documentation built on July 9, 2023, 3:09 p.m.