fda (version 2.4.7)

plot.lmWinsor: lmWinsor plot

Description

plot an lmWinsor model or list of models as line(s) with the data as points

Usage

# S3 method for lmWinsor
plot(x, n=101, lty=1:9, col=1:9,
         lwd=c(2:4, rep(3, 6)), lty.y=c('dotted', 'dashed'),
         lty.x = lty.y, col.y=1:9, col.x= col.y, lwd.y = c(1.2, 1),
         lwd.x=lwd.y, ...)

Arguments

x

an object of class 'lmWinsor', which is either a list of objects of class c('lmWinsor', 'lm') or is a single object of that double class. Each object of class c('lmWinsor', 'lm') is the result of a single 'lmWinsor' fit. If 'x' is a list, it summarizes multiple fits with different limits to the same data.

n

integer; with only one explanatory variable 'xNames' in the model, 'n' is the number of values at which to evaluate the model predictions. This is ignored if the number of explanatory variable 'xNames' in the model is different from 1.

lty, col, lwd, lty.y, lty.x, col.y, col.x, lwd.y, lwd.x

'lty', 'col' and 'lwd' are each replicated to a length matching the number of fits summarized in 'x' and used with one line for each fit in the order appearing in 'x'. The others refer to horizontal and vertical limit lines.

optional arguments for 'plot'

Value

invisible(NULL)

Details

1. One fit or several?

2. How many explanatory variables are involved in the model(s) in 'x'? If only one, then the response variable is plotted vs. that one explanatory variable. Otherwise, the response is plotted vs. predictions.

3. Plot the data.

4. Plot one line for each fit with its limits.

See Also

lmWinsor plot

Examples

Run this code
# NOT RUN {
lm.1 <- lmWinsor(y1~x1, data=anscombe)
plot(lm.1)
plot(lm.1, xlim=c(0, 15), main="other title")

# list example
lm.1. <- lmWinsor(y1~x1, data=anscombe, trim=c(0, 0.25, .4, .5)) 
plot(lm.1.)

# }

Run the code above in your browser using DataCamp Workspace