ppoints
Plotting Points for Quantile-Quantile Plots
Description
Returns a vector of points equally spaced between 0 and 1
(or unequally-spaced if weights are provided);
probabilities suitable to produce a quantile-quantile (QQ) plot.
Usage
ppoints(n, a = ifelse(n <= 10, 3/8, 1/2), weights = NULL)
Arguments
| n |
number of points desired ,
or a vector whose length is the number of points desired.
|
| a |
parameter that controls the precise placement of the plotting points,
The default value depends on the value of n,
either .375 or .5.
|
| weights |
a numeric vector, the sum of whose elements must be positive.
Missing values are not allowed. See VALUE below.
|
Details
The result is a vector of probabilities suitable for use in
probability (quantile-quantile) plots.
Plotting sort(y) against qdist(p)
(where p is this result)
gives a probability (QQ) plot of y against the distribution
for which qdist is the quantile function.
The plot should be roughly linear if y is generated
from the distribution with quantile function qdist.
Value
if no weights,
a vector of length n containing the values
(1:n - a)/(n+1-2*a).
If the length of n is less than 1 or the value of n is less than 0,
numeric(0) will return.
If weights are present the interval (0,1) is divided into
length(weights) subintervals such that the width of subinterval
i is proportional to weights[i].
The parameter a controls the spacing of returned values
from each subinterval as in the unweighted case.
References
Cleveland, William S. (1993).
Visualizing Data.
Hobart Press, Summit, New Jersey.
Differences between Spotfire Enterprise Runtime for R and Open-source R
The ppoints function in open-source R does not have the weights argument.
See Also
Examples
ppoints(5)
ppoints(10)