This question might seem silly, because I felt that it MUST be in the
mailing list archives or help files somewhere, but I simply couldn't find
it.
I want to make some simple CDF (cumulative distribution function) plots
to check whether distributions are Gaussian / normal. But in order to check
how "normal" the distribution is, I really need the y-axis to be
Gaussian as
well (meaning that the axis is not linearly scaled), so that if the
distribution is indeed normal then the plot will be a straight line.
I looked up cdf.plot (on the spsurvey package) and ecdf. They are both
fine, but neither have any way (that I can see) to alter the scale on the
y-axis so that it is not linear in the probability. How do I do this?
As long as I am asking: I also want to plot more than one distribution
or group. For instance, I want to plot 3 different factors on the same
graph, where each factor has it's own cdf line. Any ideas for this?
THanks in advance!
Mike Williamson
[[alternative HTML version deleted]]
On Mon, Dec 15, 2008 at 9:48 PM, Mike Williamson <this.is.mvw@gmail.com>wrote:> I want to make some simple CDF (cumulative distribution function) plots > to check whether distributions are Gaussian / normal.qqnorm()> As long as I am asking: I also want to plot more than one distribution > or group. For instance, I want to plot 3 different factors on the same > graph, where each factor has it's own cdf line. Any ideas for this?a=qqnorm(rnorm(100)) b=qqnorm(runif(100)) plot( a$x[order(a$x)] ,a$y[order(a$x)] ,type='l' ,ylim=c(min(c(a$y,b$y)),max(c(a$y,b$y))) ,xlim=c(min(c(a$x,b$x)),max(c(a$x,b$x))) ) lines( b$x[order(b$x)] ,b$y[order(b$x)] ,col='red' ) -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University www.thatmike.com Looking to arrange a meeting? Do so at: http://www.timetomeet.info/with/mike/ ~ Certainty is folly... I think. ~ [[alternative HTML version deleted]]