qualityTools (version 1.53)

normalPlot: Normal plot

Description

function to generate a normal plot of the factor effects for an object of class facDesign.

Usage

normalPlot(fdo, threeWay = FALSE, na.last = NA, alpha = 0.05, response = NULL,
           sig.col = c("red1", "red2", "red3"), main, ylim, xlim, xlab, ylab,
           pch, col, border = "red", ...)

Arguments

fdo

object of class facDesign.

threeWay

TRUE/FALSE plot three-way or higher interactions. By default threeWay is set to FALSE.

na.last

By default set to NA.

alpha

alpha for marking interactions.

response

response variable. If the response data frame of fdo consists of more then one responses, this variable can be used to choose just one column of the response data frame. response needs to be an object of class character with length of

sig.col

vector - colors for marking significant interactions. By default sig.col is set to c("red1", "red2", "red3").

main

graphical parameter. A main title for the plot, see also title.

ylim

graphical parameter. The y limits of the plot.

xlim

graphical parameter. The x limits (x1, x2) of the plot. Note that x1 > x2 is allowed and leads to a reversed axis.

xlab

graphical parameter. A label for the x axis, defaults to a description of x.

ylab

graphical parameter. A label for the y axis, defaults to a description of y.

pch

graphical parameter. Vector containing numerical values or single characters giving plotting points for the different factors. See points for possible values and their interpretation. Note that only integers

col

graphical parameter. Single numerical value or character string giving the color for the points.

border

graphical parameter. Single numerical value or character string giving the color of the border line.

...

further graphical parameters see par.

Value

• NULL

Details

If the given facDesign object fdo contains replicates this function will deliver a normal plot i.e.: effects divided by the standard deviation (t-value) will be plotted against an appropriate probability scaling (see: ppoints). If the given facDesign object fdo contains no replications the standard error can not be calculated. In that case the function will deliver an effect plot. i.e.: the effects will be plotted against an appropriate probability scaling. (see: ppoints).

See Also

facDesign paretoPlot http://www.r-qualitytools.org/Improve.html

Examples

#factorial design
fdo = facDesign(k=3, replicates = 2)  
#seed for random numbers
set.seed(123) 
#random numbers
y = rnorm(nrow(fdo)) 
#set the response 
response(fdo) = y 
#create a normal plot
normalPlot(fdo)