[Traminer-users] precision in seqdef

Nathan Green n8thangreen at yahoo.co.uk
Fri Jun 24 10:41:37 CEST 2011


I'm using the following SPELL format data.
The problem is that it appears the precision of the time data is too high for the seqdef function
when I round to 3 d.p. its fine.
what would you suggest i should do in this case?
I can premultiply by something large and then divide afterwards?
 
thanks for the help.
 
n
 
 
id index from to status
1 1 1 24.00000 179.66694 1
2 1 2 179.66694 181.33361 2
3 1 3 181.33361 191.33361 4
4 1 4 191.33361 201.33361 5
5 2 1 24.00000 376.00667 1
6 2 2 376.00667 400.00667 2
7 2 3 400.00667 410.00667 4
8 2 4 410.00667 420.00667 5
9 3 1 24.00000 40.00667 1
10 3 2 40.00667 56.00667 2
11 3 3 56.00667 58.00667 4
12 3 4 58.00667 68.00667 5
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/traminer-users/attachments/20110624/ab7894eb/attachment.htm>


More information about the Traminer-users mailing list