Predict textmodel_wordscores
# S3 method for textmodel_wordscores
predict(object, newdata = NULL,
se.fit = FALSE, interval = c("none", "confidence"), level = 0.95,
rescaling = c("none", "lbg", "mv"), include_reftexts = TRUE, ...)a fitted Wordscores textmodel
dfm on which prediction should be made
if TRUE, return standard errors as well
type of confidence interval calculation
tolerance/confidence level for intervals
"none" for "raw" scores; "lbg" for LBG (2003)
rescaling; or "mv" for the rescaling proposed by Martin and Vanberg
(2007). See References.
if FALSE, reference texts are removed from the prediction
not used
textmodel_wordscores() returns a list that is also classed as a
textmodel_wordscores object, containing the following elements:
the scores computed for each word in the training set (\(S_{wd}\) from Laver, Benoit and Garry 2003)
either linear or logit, according to the value of scale
the dfm on which the wordscores model was called
the vector of document reference values
the function call that fitted the model
predict.textmodel_wordscores() returns a named vector of predicted document scores ("text scores" S_{vd} in LBG 2003), or a named list if se.fit = TRUE consisting of the predicted scores ($fit) and the associated standard errors ($se.fit). When interval = "confidence", the predicted values will be a matrix. This behaviour matches that of predict.lm.