[R] calculating time interval distributions

2011-11-16 Thread Grant Gillis
Dear List, I have data on a approximately 100 individuals visiting a a central logging station over a 1000 times. I would like to be able to calculate the distribution of inter-visit time intervals for all possible pairs am stuck on how to code for this. Single pairs are not a problem but

Re: [R] calculating time interval distributions

2011-11-16 Thread Bert Gunter
Your question seems rather poorly specified to me, and you may have to clarify it further, but I'll make a stab at it. IF one can assume that: 1. a must occur before b, b before c etc. AND 2. sec is always increasing THEN z - tapply(sec, ind, FUN = min) gives the first appearance sec value