The following describes all estimation function types used in matrixpls package.
A function for calculating indicator weights using the data covariance matrix
S, a model specification model, and a weight pattern W.model. Returns
a weight matrix W. The default is weightFun.pls
A function for estimating the model parameters using
the data covariance matrix S, model specification model,
and weight matrix W. Returns a named vector of parameter estimates.
The default is parameterEstim.separate
A function for estimating the parameters of one model matrix using
the data covariance matrix S, a model matrix modelMatrix, and a weight matrix
W. Disattenuated composite correlation matrix C and indicator composite
covariance matrix IC are optional. Returns matrix of parameter estimates.
The default is estimator.ols
A function for resolving weight sign ambiguity based on the data covariance matrix
S and a weight matrix W. Returns
a weight matrix W. See weightSign
for details.
A function for calculating outer weights using the data covariance matrix
S, a weight matrix W, an inner weight matrix E,
and a weight pattern W.model. Returns
a weight matrix W. See outerEstim.
A function for calculating inner weights using the data covariance matrix
S, a weight matrix W, and an inner model matrix inner.mod. Returns
an inner weight matrix E. The default is innerEstim.path.
A function that takes the old Wold and new weight Wold matrices and
returns a scalar that is compared against tol to check for convergence. The default
is convCheck.absolute.
A function for calculating value for an optimization criterion based on a
matrixpls result object. Returns a scalar. The default is optimCrit.maximizeInnerR2.
A function for calculating reliability estimates based on the
data covariance matrix S, factor loading matrix loadings, and a weight matrix W.
Returns a vector of reliability estimates. The default is
reliabilityEstim.weightLoadingProduct