predict.glmrob: Predict Method for Robust GLM ("glmrob") Fits

View source: R/glmrobPredict.R

predict.glmrobR Documentation

Predict Method for Robust GLM ("glmrob") Fits


Obtains predictions and optionally estimates standard errors of those predictions from a fitted robust generalized linear model (GLM) object.


## S3 method for class 'glmrob'
predict(object, newdata = NULL,
       type = c("link", "response", "terms"), = FALSE,
       dispersion = NULL, terms = NULL, na.action = na.pass, ...)



a fitted object of class inheriting from "glmrob".


optionally, a data frame in which to look for variables with which to predict. If omitted, the fitted linear predictors are used.


the type of prediction required. The default is on the scale of the linear predictors; the alternative "response" is on the scale of the response variable. Thus for a default binomial model the default predictions are of log-odds (probabilities on logit scale) and type = "response" gives the predicted probabilities. The "terms" option returns a matrix giving the fitted values of each term in the model formula on the linear predictor scale.

The value of this argument can be abbreviated.

logical switch indicating if standard errors are required.


the dispersion of the GLM fit to be assumed in computing the standard errors. If omitted, that returned by summary applied to the object is used.


with type="terms" by default all terms are returned. A character vector specifies which terms are to be returned


function determining what should be done with missing values in newdata. The default is to predict NA.


optional further arguments, currently simply passed to predict.lmrob().


If se = FALSE, a vector or matrix of predictions. If se = TRUE, a list with components



Estimated standard errors


A scalar giving the square root of the dispersion used in computing the standard errors.


Andreas Ruckstuhl

See Also

glmrob() to fit these robust GLM models, residuals.glmrob() and other methods; predict.lm(), the method used for a non-robust fit.


## simplistic testing & training: <- sample(24, 20)
fm1 <- glmrob(cbind(success, total-success) ~ logdose + block,
              family = binomial, data = carrots, subset =
predict(fm1, carrots[, ]) # --> numeric vector
predict(fm1, carrots[, ],
        type="response", se = TRUE)# -> a list

Vfit <- glmrob(Y ~ log(Volume) + log(Rate), family=binomial, data=vaso)
newd <- expand.grid(Volume = (V. <- seq(.5, 4, by = 0.5)),
                    Rate   = (R. <- seq(.25,4, by = 0.25)))
p <- predict(Vfit, newd)
filled.contour(V., R., matrix(p, length(V.), length(R.)),
      main = "predict(glmrob(., data=vaso))", xlab="Volume", ylab="Rate")

robustbase documentation built on April 3, 2022, 1:05 a.m.