condvis (version 0.2-2)

plotxs: Expectation plot

Description

Plot depicting a section through a fitted model surface (or a conditional expectation). Not designed to be called by the user.

Usage

plotxs(xs, y, xc.cond, model, model.colour = NULL, model.lwd = NULL,
       model.lty = NULL, model.name = NULL, yhat = NULL, mar = NULL, 
       data.colour = NULL, data.order = NULL, view3d = FALSE, theta3d = 45,
       phi3d = 20)

Arguments

xs

a data.frame with one or two columns.

y

a data.frame with one column.

xc.cond

a data.frame with one row.

model

a model object, or a list of model objects.

model.colour, model.lwd, model.lty

graphical parameters passed to col, lwd and lty respectively in a call to points (to differentiate between models).

model.name

names for the models in model, used for the legend.

yhat

a list of fitted values for the models in model

mar

margins for plotting.

data.colour

vector of colours for plotting the raw data (xs and/or y)

data.order

integer vector giving the order in which to plot raw data

view3d

logical; if TRUE, a 3-d perspective plot is produced if possible

theta3d

horizontal rotation of perspective plot

phi3d

vertical rotation of perspective plot

Value

• A list containing:
• xs
• y
• xc.cond
• model
• model.colour
• model.lwd
• model.lty
• model.name
• yhat
• mar
• data.colour
• data.order
• view3d
• theta3d
• phi3d

See Also

ceplot, plotxc.

Examples

## expected value of 'mpg' conditional on fitted model and predictor values in 
## first observation of mtcars, evaluated for a range of 'wt' values.

data(mtcars)
model <- lm(mpg ~ ., data = mtcars)
plotxs(xs = mtcars[, "wt", drop = FALSE], 
    y = mtcars[, "mpg", drop = FALSE], 
    xc.cond = mtcars[1, ], 
    model = list(model))