[Traminer-users] Problem creating a sequence object from spell to sts

Gilbert Ritschard Gilbert.Ritschard at unige.ch
Fri Jun 8 09:57:06 CEST 2012


Hi Anton,

The summaries that your report are NOT those of your begin and end 
variables. Seems you are computed ages from year data and use the ages 
as begin and end variable. The solution consist in just adding 1 to the 
P_AGEDEB and P_AGEFIN variables. Then it is just a matter of 
interpretation: the resulting ages should be seen as ages at the and of 
the period.

Hope this helps.

Gilbert

On 07-Jun-12 16:40, Anton Perdoncin wrote:
> Hello,
>
> This might be a stupid question, but an error message pops up in the 
> console when I try to create a sequence object :
> Erreur :  [!] found one or more spell with starting time<  1
>
> Here is the code:
> LA.sts <- seqformat(trajpro, id="IDENT", begin="P_AGDEB", 
> end="P_AGFIN", status="traj.cs", from="SPELL", to="STS", process=FALSE)
>
> And the summary of the begin and end variables;
> summary(trajpro$P_ANDEB)
>     Min. 1st Qu.  Median    Mean 3rd Qu.    Max.
>     1948    1976    1989    1987    2000    2009
> >  summary(trajpro$P_ANFIN)
>     Min. 1st Qu.  Median    Mean 3rd Qu.    Max.
>     1948    1992    2003    1999    2008    2009
>
>
> It seems that there is no missing value.
>
> The dataset is Ined Trajectoires et Origines (TeO). I suspect that the problem comes from the way trajectories where coded: when one individual spends less than one year in a particular state, begin and end years are the same. So spell=0...
>
> Am I right? Does it imply that I have to recode the whole dataset?
>
> Thanks in advance for your help.
>
> Regards,
>
> Anton Perdoncin
>
>
>
>
> _______________________________________________
> 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


More information about the Traminer-users mailing list