gap (version 1.2.3-1)

qqunif: Q-Q plot for uniformly distributed random variable

Description

This function produces Q-Q plot for a random variable following uniform distribution with or without using log-scale. Note that the log-scale is by default for type "exp", which is a plot based on exponential order statistics. This appears to be more appropriate than the commonly used procedure whereby the expected value of uniform order statistics is directly log-transformed.

Usage

qqunif(u,type="unif",logscale=TRUE,base=10,
       col=palette()[4],lcol=palette()[2],ci=FALSE,alpha=0.05,...)

Arguments

u

a vector of uniformly distributed random variables

type

string option to specify distribution: "unif"=uniform, "exp"=exponential

logscale

to use logscale

base

the base of the log function

col

color for points

lcol

color for the diagonal line

ci

logical option to show confidence interval

alpha

1-confidence level, e.g., 0.05

...

other options as appropriae for the qqplot function

Value

The returned value is a list with components of a qqplot:

x

expected value for uniform order statistics or its -log(,base) counterpart

y

observed value or its -log(,base) counterpart

References

Balakrishnan N, Nevzorov VB. A Primer on Statistical Distributions. Wiley 2003.

Casella G, Berger RL. Statistical Inference, Second Edition. Duxbury 2002.

Davison AC. Statistical Models. Cambridge University Press 2003.

See Also

qqfun

Examples

# NOT RUN {
# Q-Q Plot for 1000 U(0,1) r.v., marking those <= 1e-5
u_obs <- runif(1000)
r <- qqunif(u_obs,pch=21,bg="blue",bty="n")
u_exp <- r$y
hits <- u_exp >= 2.30103
points(r$x[hits],u_exp[hits],pch=21,bg="green")
legend("topleft",sprintf("GC.lambda=%.4f",gc.lambda(u_obs)))
# }