From mummyra at gmail.com Mon Nov 14 02:59:36 2016 From: mummyra at gmail.com (Mira Cho) Date: Mon, 14 Nov 2016 10:59:36 +0900 Subject: [Traminer-users] how to get representative sequence by using multichannel method Message-ID: Dear tramier-users Can I get representative sequence by using multichannel method? I am trying to analyze couple's time schedules. If I use multichannel distance matrix (getting by "seqdismc" using list: male & female sequences), "seqrep" would resurns it? I just guess it will be differently affected according to criterion sorting representative candidate list such as "freq", "dist"... "dist" is about centrality having the smallest sum of the distances to all othere sequences, Then can I think it will be affected just by distance matrix for returning representative? best, Mira Cho Mira Ph.D.candidate Social welfare department Seoul national university South Korea -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gilbert.Ritschard at unige.ch Fri Nov 25 09:18:03 2016 From: Gilbert.Ritschard at unige.ch (Gilbert Ritschard) Date: Fri, 25 Nov 2016 08:18:03 +0000 Subject: [Traminer-users] how to get representative sequence by using multichannel method In-Reply-To: References: Message-ID: <013beaf7fe044f40a4cb7f58bcf5a311@wuniex13-2.isis.unige.ch> Hi Mira, Sorry for this late answer. Currently you cannot get representative multichannel sequences. The algorithm used for searching for the representative sequence(s) requires only the dissimilarity matrix. It should therefore be applicable to a dissimilarity matrix of multichannel sequences. However, the seqrep function that implements it requires also a state sequence object that is used for printing the results. It does not know how to handle the list of the multichannel state sequences. Actually, the print and plot function for the object returned by seqrep work only with single channel sequences. A solution could be to transform the multichannel sequences into combined single sequences and pass the resulting sequence object to seqrep together with the dissimilarity matrix computed using seqdistmc. Best. Gilbert From: traminer-users-bounces at lists.r-forge.r-project.org [mailto:traminer-users-bounces at lists.r-forge.r-project.org] On Behalf Of Mira Cho Sent: lundi 14 novembre 2016 03:00 To: traminer-users at lists.r-forge.r-project.org Subject: [Traminer-users] how to get representative sequence by using multichannel method Dear tramier-users Can I get representative sequence by using multichannel method? I am trying to analyze couple's time schedules. If I use multichannel distance matrix (getting by "seqdismc" using list: male & female sequences), "seqrep" would resurns it? I just guess it will be differently affected according to criterion sorting representative candidate list such as "freq", "dist"... "dist" is about centrality having the smallest sum of the distances to all othere sequences, Then can I think it will be affected just by distance matrix for returning representative? best, Mira Cho Mira Ph.D.candidate Social welfare department Seoul national university South Korea -------------- next part -------------- An HTML attachment was scrubbed... URL: From Gilbert.Ritschard at unige.ch Mon Nov 28 15:12:35 2016 From: Gilbert.Ritschard at unige.ch (Gilbert Ritschard) Date: Mon, 28 Nov 2016 14:12:35 +0000 Subject: [Traminer-users] how to get representative sequence by using multichannel method In-Reply-To: <013beaf7fe044f40a4cb7f58bcf5a311@wuniex13-2.isis.unige.ch> References: <013beaf7fe044f40a4cb7f58bcf5a311@wuniex13-2.isis.unige.ch> Message-ID: <7e3070170b764949a147f0dc742f2281@wuniex13-2.isis.unige.ch> Hi Mira, I forgot there is a TraMineR dissrep function that searches for representatives from any dissimilarity matrix. You could use it to get representatives of multichannel sequences. However, there is no plot function for the resulting outcome. Best. Gilbert From: traminer-users-bounces at lists.r-forge.r-project.org [mailto:traminer-users-bounces at lists.r-forge.r-project.org] On Behalf Of Gilbert Ritschard Sent: vendredi 25 novembre 2016 09:18 To: Users questions Subject: Re: [Traminer-users] how to get representative sequence by using multichannel method Hi Mira, Sorry for this late answer. Currently you cannot get representative multichannel sequences. The algorithm used for searching for the representative sequence(s) requires only the dissimilarity matrix. It should therefore be applicable to a dissimilarity matrix of multichannel sequences. However, the seqrep function that implements it requires also a state sequence object that is used for printing the results. It does not know how to handle the list of the multichannel state sequences. Actually, the print and plot function for the object returned by seqrep work only with single channel sequences. A solution could be to transform the multichannel sequences into combined single sequences and pass the resulting sequence object to seqrep together with the dissimilarity matrix computed using seqdistmc. Best. Gilbert From: traminer-users-bounces at lists.r-forge.r-project.org [mailto:traminer-users-bounces at lists.r-forge.r-project.org] On Behalf Of Mira Cho Sent: lundi 14 novembre 2016 03:00 To: traminer-users at lists.r-forge.r-project.org Subject: [Traminer-users] how to get representative sequence by using multichannel method Dear tramier-users Can I get representative sequence by using multichannel method? I am trying to analyze couple's time schedules. If I use multichannel distance matrix (getting by "seqdismc" using list: male & female sequences), "seqrep" would resurns it? I just guess it will be differently affected according to criterion sorting representative candidate list such as "freq", "dist"... "dist" is about centrality having the smallest sum of the distances to all othere sequences, Then can I think it will be affected just by distance matrix for returning representative? best, Mira Cho Mira Ph.D.candidate Social welfare department Seoul national university South Korea -------------- next part -------------- An HTML attachment was scrubbed... URL: From reyno113 at purdue.edu Wed Nov 30 16:51:45 2016 From: reyno113 at purdue.edu (Reynolds, Jeremy E) Date: Wed, 30 Nov 2016 15:51:45 +0000 Subject: [Traminer-users] remove titles from subplots Message-ID: <0e3d85e8d3be4178a30d03a7149524a8@wppexc08.purdue.lcl> Hi, Does anyone know how to remove the subtitles from the subplots that are created with the group option in the seqdplot command? I didn't see any mention of them in the par help file. Perhaps I have to do this by somehow removing the labels from the grouping variable? I have pasted code for an example graph below. Thanks, Jeremy library(TraMineR) library(WeightedCluster) data(mvad) mvad.alphabet <- c("employment", "FE", "HE", "joblessness", "school", "training") mvad.labels <- c("Employment", "Further Education", "Higher Education", "Joblessness", "School", "Training") mvad.scodes <- c("EM", "FE", "HE", "JL", "SC", "TR") mvadseq <- seqdef(mvad[, 17:86], alphabet = mvad.alphabet, states = mvad.scodes, labels = mvad.labels, weights = mvad$weight, xtstep = 6) #Defining the custom cost matrix subm.custom <- matrix( c(0, 1, 1, 2, 1, 1, 1, 0, 1, 2, 1, 2, 1, 1, 0, 3, 1, 2, 2, 2, 3, 0, 3, 1, 1, 1, 1, 3, 0, 2, 1, 2, 2, 1, 2, 0), nrow = 6, ncol = 6, byrow = TRUE) mvaddist1.0 <- seqdist(mvadseq, method = "OM", indel = 1.0, sm = subm.custom) pamclust1.0 <- wcKMedRange(mvaddist1.0, kvals = 2:10, weights = mvad$weight) seqdplot(mvadseq, group = pamclust1.0$clustering$cluster4, border=NA) Dr. Jeremy Reynolds Professor 309 Stone Hall Department of Sociology 700 W. State Street Purdue University West Lafayette, IN 47907 Phone: (765) 496-3348 https://www.cla.purdue.edu/sociology/directory/index.aspx?p=Jeremy_Reynolds -------------- next part -------------- An HTML attachment was scrubbed... URL: