Known sampling covariance marker

meta_V() marks known sampling variance or covariance in a formula. It is designed for meta-analysis and other regression problems where part of the observation covariance is known in advance.

Usage

meta_V(V)

Arguments

V

A column name, vector, diagonal matrix, block-diagonal matrix, or full covariance matrix. Diagonal/vector V represents independent known sampling variances. A matrix represents the known covariance among rows. meta_V() matches V to the model frame by row position after ordinary drmTMB() row handling; it does not reorder a covariance matrix by row names or dimnames. Put the rows of data and the rows and columns of V in the same order before fitting.

Value

A formula marker; never evaluated by users.

Examples

bf(yi ~ moderator + meta_V(V = vi), sigma ~ moderator)
#> <drm_formula>
#> yi ~ moderator + meta_V(V = vi)
#> sigma ~ moderator