memisc (version 0.97)

prediction.frame: Produce a Data Frame of Predictions and Independent Variables

Description

prediction.frame produces a data frame that contains the independent variables of a model together with model generated predictions.

Usage

## S3 method for class 'default':
prediction.frame(object,newdata=NULL,...,
          residuals=c("none", "deviance", "pearson", "working",
                      "standardized", "studentized"))

Arguments

object

a model object from which predictions are generated.

newdata

an optional data frame for out-of-sample predictions.

...

further arguments passed to predict.

residuals

a character vector that specifies residuals that are added to the resulting data frame.

Value

• A data frame.

Examples

lm1 <- lm(sr ~ pop15 + pop75 + dpi + ddpi, data = LifeCycleSavings)
lm2 <- lm(sr ~ pop15 + pop75 + dpi + ddpi + pop15*dpi, data = LifeCycleSavings)

str(prediction.frame(lm1))
str(prediction.frame(lm1, se.fit=TRUE))
str(prediction.frame(lm1, interval="prediction"))
str(prediction.frame(lm1, type="terms"))
str(prediction.frame(lm1,
                  se.fit=TRUE,
                  type="terms"))

str(prediction.frame(lm1,
                  se.fit=TRUE,
                  type="terms",
                  residuals="working"))

str(prediction.frame(lm1,
                  se.fit=TRUE,
                  type="terms",
                  residuals="deviance"))

str(prediction.frame(lm2,
                  se.fit=TRUE,
                  type="terms",
                  residuals="standardized"))

berkeley <- Aggregate(Table(Admit,Freq)~.,data=UCBAdmissions)
berk2 <- glm(cbind(Admitted,Rejected)~Gender+Dept,data=berkeley,family="binomial")

str(prediction.frame(berk2,
                  se.fit=TRUE,
                  type="terms",
                  residuals="studentized"))