predict.MxModel {OpenMx}

predict method for MxModel objects

Description

predict method for MxModel objects

Usage

## S3 method for class 'MxModel'
predict(
  object,
  newdata = NULL,
  interval = c("none", "confidence", "prediction"),
  method = c("ML", "WeightedML", "Regression", "Kalman"),
  level = 0.95,
  type = c("latent", "observed"),
  ...
)

Arguments

object

an MxModel object from which predictions are desired

newdata

an optional data.frame object. See details.

interval

character indicating what kind of intervals are desired. 'none' gives no intervals, 'confidence', gives confidence intervals, 'prediction' gives prediction intervals.

method

character the method used to create the predictions. See details.

level

the confidence or predictions level, ignored if not using intervals

type

character the type of thing you want predicted: latent variables or manifest variables.

...

further named arguments

Details

The newdata argument is either a data.frame or MxData object. In the latter case is replaces the data in the top level model. In the former case, it is passed as the observed argument of mxData with type='raw' and must accept the same further arguments as the data in the model passed in the object argument.

The available methods for prediction are 'ML', 'WeightedML', 'Regression', and 'Kalman'. See the help page for mxFactorScores for details on the first three of these. The 'Kalman' method uses the Kalman filter to create predictions for state space models.