Inverse cumulative probability
On Tue, 14 Feb 2006, Jan Danielsson wrote:
Hello all, (First of all, I'd like to thank all who replied to my previous question. I have never encountered such a helpful community before. Thanks for making a R so welcoming.) To calculate a quantile for normal distributions, one simply uses qnorm(1-a).
If a is small, it is better to use qnorm(a, lower.tail=FALSE).
But if I would want to do the same for a t-test function, how would I go about doing that? Is there a simple way to do it? (Yes, I could look in a table, but it's the procedure I'm looking for, not the values :-)
qt(a, nu, lower.tail=FALSE), assuming by `t-test function' you mean Student's t distribution on nu degrees of freedom (which might be n-1 or n-2 for a t test). For more on this see `An Introduction to R', specifically section 8.1 in the HTML version I just looked at.
Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595