[Traminer-users] creating multicharacter state sequence objects &weights

Matthias Studer Matthias.Studer at unige.ch
Mon Feb 13 11:09:57 CET 2012


Dear Daïga,

Here are some answers to your questions.

1) The function "interaction" from the "base" library can be used to 
create a new variable. For instance, to create a new variable called 
"state1", you may use the following code:

mydata$state1 <- interaction(mydata$A1, mydata$B1, mydata$C1, mydata$D1, 
mydata$E1)

Doing it for each of the five time t, you can get your sequence data. 
For more information on the "interaction" function, you may run:

help(interaction)

2) sampling weights can be used with the weights argument of the seqdef 
function (see help(seqdef)). By default, weights will be used in all 
subsequent call to the different seq type functions. More information on 
this are given in the reference below (see section 4)

Gabadinho, A., Ritschard, G., Müller, N.S. & Studer, M. (2011), 
Analyzing and visualizing state sequences in R with TraMineR, Journal of 
Statistical Software. Vol. 40(4), pp. 1-37.

Hope this helps

Matthias Studer


Le 09.02.2012 09:40, Kamerade Daiga a écrit :
> Dear All,
>
> firstly, hello to everyone- I am new to this list lured here by 
> analytical opportunities offered by TraMineR which are excellently 
> presented in the User Guide! I am using TraMineR to analyse membership 
> histories in voluntary sector organisations.
>
> Two questions:
>
> 1) Could anyone, please, help me to figure out how can I create 
> multi-character state sequence objects. Currently I have five 
> variables A, B, C,D,E measured five times, each having two levels: yes 
> (1) &no (0), so the data frame contains 25 colums with states: A1,B1, 
> C1, D1, E1, A2, B2, C2, D2, E2, A3....etc E5. I want to create 
> sequences where each state is represented by all five variables at 
> time point t., e.g. something like this: 01111-10000-11111, i.e at 
> time t this persona was not in A but was in BCDE; at time t+1 he was 
> in A but not in other etc.
> 2) how can sampling weights be incorporated into different seq type 
> functions?
>
> thank you very much in advance!
>
> best wishes
> Daiga
>
>
>
>
> _____________________________________________________________________
> *Daiga Kamerade, Ph.D. (Cantab), PGCE HEPR, FHEA*
> Lecturer in Organisational Behaviour and HRM, Salford Business School
> VC Early Career Research Scholar
> Room 327, Maxwell Building, University of Salford, Salford, 
> Manchester, M5 4WT, UK
> Tel. + 44 (0)161 295 5318
> Research 
> profile:http://www.seek.salford.ac.uk/profiles/DAIGA%20KAMERADE.jsp
>
>
>
> _______________________________________________
> 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/20120213/03aa5b95/attachment.html>


More information about the Traminer-users mailing list