Re: [R] p-val issue for ranked two-group test

2011-10-20 Thread Joshua Wiley
On Thu, Oct 20, 2011 at 3:42 PM, peter dalgaard wrote: > > On Oct 20, 2011, at 23:48 , Joshua Wiley wrote: > >> Hi, >> >> It looks like you are trying to manually bootstrap. > > Nope. It's a manually performed approximate Wilcoxon test. Which is fair > enough if the object is to learn something.

Re: [R] p-val issue for ranked two-group test

2011-10-20 Thread peter dalgaard
On Oct 20, 2011, at 23:48 , Joshua Wiley wrote: > Hi, > > It looks like you are trying to manually bootstrap. Nope. It's a manually performed approximate Wilcoxon test. Which is fair enough if the object is to learn something. (Notice, however, that the ExactRankTests package eats this sort

Re: [R] p-val issue for ranked two-group test

2011-10-20 Thread Joshua Wiley
Hi, It looks like you are trying to manually bootstrap. Take a look at: require(boot) ?boot as an added advantage of using boot instead of trying to do it manually, you can easily parallelize. In fact, if you are using one of the pre-release versions of 2.14.0, the new parallel package is incl

[R] p-val issue for ranked two-group test

2011-10-20 Thread Laurel Klein Serieys
Hi- I'm wondering if anyone can help me with my code. I'm coming up dry when I try to get a p-value from the following code. If I make a histogram of my resampled distribution, I find the difference between by groups to be significant. I've ranked the data since I have outliers in one o