[Traminer-users] association rules from event sequences

[Nicolas Müller]

Dear Hugo,

The functions needed to get the hazard ratios and the p-values from the 
association rules are not available yet in TraMineR.
However, I will make them available in the TraMineRextras package during 
this week-end, along with some explanations on how to use them.
They are not very optimized yet, and I'd be glad if someone was 
interested in improving the code, as I don't have enough free time to do 
it at the moment.

Best regards.

Nicolas

Le 20.03.2012 12:02, Hugo Varet a écrit :
> Dear all,
>
> I have a new question about the analysis of event sequences. I would 
> like extracting association rules from my data as Nicolas Müller did 
> in his paper entitled "Extraction de règles d'association séquentielle 
> à l'aide de modèles semi-paramétriques à risques proportionnels". His 
> way of presenting the results (hazard ratio with p-value from a Cox 
> model) is very interesting and understandable for a large public. I 
> succeeded in extracting association rules with the following R code :
> freq=seqefsub(data.seqe, pMinSupport = 0.05)
> rules <- TraMineR:::seqerules(freq)
> rules[order(rules$Lift, decreasing = TRUE)[1:10],]
> but I do not know how to get HR and p-values (maybe I do not use the 
> good fonction). Nicolas Müller wrote in his paper that his algorithm 
> would be implemented in the TraMineR package. Maybe the fonction I am 
> looking for is not available yet ?
>
> Best regards,
>
> Hugo
>
>
> _______________________________________________
> Traminer-users mailing list
> Traminer-users at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/traminer-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/traminer-users/attachments/20120324/6d9ec60d/attachment.html>