lavaSearch2 (version 1.5.6)

conditionalMoment: Prepare the Computation of score2

Description

Compute the conditional mean and variance, and their first and second derivative regarding the model parameters.

Usage

conditionalMoment(object, ...)
# S3 method for lm
conditionalMoment(
  object,
  data,
  param,
  name.endogenous,
  first.order,
  second.order,
  ...
)
# S3 method for gls
conditionalMoment(
  object,
  data,
  formula,
  param,
  attr.param,
  ref.group,
  first.order,
  second.order,
  index.Omega,
  vec.OmegaMat,
  cluster,
  n.cluster,
  name.endogenous,
  n.endogenous,
  ...
)
# S3 method for lme
conditionalMoment(object, attr.param, ...)
# S3 method for lvm
conditionalMoment(
  object,
  data,
  first.order,
  second.order,
  name.endogenous,
  name.latent,
  ...
)
# S3 method for lvmfit
conditionalMoment(object, data, param, first.order, second.order, usefit, ...)

Arguments

object, x

a latent variable model.

...

[internal] only used by the generic method or by the <- methods.

data

[data.frame] data set.

param, p

[numeric vector] the fitted coefficients.

name.endogenous

[character vector, optional] name of the endogenous variables

second.order

[logical] should the terms relative to the third derivative of the likelihood be be pre-computed?

formula

[formula] two-sided linear formula.

attr.param

[character vector] the type of each coefficient (e.g. mean or variance coefficient).

ref.group

[character vector] the levels of the variable defining the variance component in a generic covariance matrix.

cluster

[integer vector] the grouping variable relative to which the observations are iid.

n.cluster

[integer >0] the number of i.i.d. observations.

n.endogenous

[integer >0] the number of outcomes.

name.latent

[character vector, optional] name of the latent variables

usefit, value

[logical] If TRUE the coefficients estimated by the model are used to pre-compute quantities. Only for lvmfit objects.

Details

For lvmfit objects, there are two levels of pre-computation:

• a basic one that do no involve the model coefficient (conditionalMoment.lvm).

• an advanced one that require the model coefficients (conditionalMoment.lvmfit).

Examples

m <- lvm(Y1~eta,Y2~eta,Y3~eta)
latent(m) <- ~eta

d <- lava::sim(m,1e2)
e <- estimate(m, d)

## basic pre-computation
res1 <- conditionalMoment(e, data = d,
                         first.order = FALSE, second.order = FALSE,
                         name.endogenous = endogenous(e),
                         name.latent = latent(e), usefit = FALSE)
res1$skeleton$Sigma

## full pre-computation
res2 <- conditionalMoment(e, param = coef(e), data = d,
                         first.order = FALSE, second.order = FALSE,
                         name.endogenous = endogenous(e),
                         name.latent = latent(e), usefit = TRUE
)
res2$value$Sigma