Related questions to this have been asked before, but I have tried all
options they gave me unsuccessfully (do.call and unlist).
I start with three lists of summary statistics, 100 elements each, which I
bind together:
statslist <- as.data.frame(cbind (means, vars, mcrs))
I then take 100 samples of this data frame of varying lengths:
stats <- lapply (1:100, function (dummy) {
statslist[sample(nrow(statslist), (sample (10:20, 1, replace = TRUE)),
replace = TRUE),]})
It returns basically what I want:
> stats[[i]]
means vars mcrs
71 1.81 3.832222 2.92725
9 2.56 8.127677 4.734874
91 3.44 9.66303 5.24902
68 0.14 0.1216162 0.008686869
except that:
> is.list(stats[[i]]$means)
[1] TRUE
I don't want this to be a list, because I want to do regressions using the
variable which are in columns, which apparently R won't do if they are in
lists.
I have tried every possible combination of working with do.call, rbind,
as.data.frame, etc to get this into matrix form to no avail.
I have also tried using sapply instead of lapply, but that returns vectors
which again, R doesn't want to do regression on.
Any tips would be very much appreciated, been going around in circles for a
while here.
Edward Waters
PhD Student UNSW
--
View this message in context:
http://n4.nabble.com/Lists-into-matrices-within-lists-again-tp1565433p1565433.html
Sent from the R help mailing list archive at Nabble.com.
On Feb 22, 2010, at 8:27 PM, ewaters wrote:> > Related questions to this have been asked before, but I have tried all > options they gave me unsuccessfully (do.call and unlist). > > I start with three lists of summary statistics, 100 elements each, > which I > bind together: >None of this represents data that are too large to include in an unambiguous fashion with dump or dput. Why do youy persist in asking questions whare there remain ambiguities?> statslist <- as.data.frame(cbind (means, vars, mcrs)) > > I then take 100 samples of this data frame of varying lengths: > > stats <- lapply (1:100, function (dummy) { > statslist[sample(nrow(statslist), (sample (10:20, 1, replace = TRUE)), > replace = TRUE),]})So what are readers supposed to do? Create similar structures and work on them?> > It returns basically what I want: > >> stats[[i]] > means vars mcrs > 71 1.81 3.832222 2.92725 > 9 2.56 8.127677 4.734874 > 91 3.44 9.66303 5.24902 > 68 0.14 0.1216162 0.008686869 > > except that: > >> is.list(stats[[i]]$means) > [1] TRUESo lapply on a list returns a list? Where is the surprise?> > I don't want this to be a list, because I want to do regressions > using the > variable which are in columns, which apparently R won't do if they > are in > lists. > > I have tried every possible combination of working with do.call, > rbind, > as.data.frame, etc to get this into matrix form to no avail. > > I have also tried using sapply instead of lapply, but that returns > vectors > which again, R doesn't want to do regression on.???? R does not want to do regressions on vectors??? How can that be?> > Any tips would be very much appreciated, been going around in > circles for a > while here. >Reproducible data would answer these questions and prevent this "circular" process that _you_ are responsible for allowing to continue. -- David.> Edward Waters > PhD Student UNSW > > > -- > View this message in context: http://n4.nabble.com/Lists-into-matrices-within-lists-again-tp1565433p1565433.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
Tena koe Edward It is difficult to know the best approach from the information supplied, but using unlist() at the appropriate place in your code, maybe unlist(stats[[i]]$means), will probably get you where you want to be. HTH ..... Peter Alspach> -----Original Message----- > From: r-help-bounces at r-project.org > [mailto:r-help-bounces at r-project.org] On Behalf Of ewaters > Sent: Tuesday, 23 February 2010 3:27 p.m. > To: r-help at r-project.org > Subject: [R] Lists into matrices within lists...again > > > Related questions to this have been asked before, but I have > tried all options they gave me unsuccessfully (do.call and unlist). > > I start with three lists of summary statistics, 100 elements > each, which I bind together: > > statslist <- as.data.frame(cbind (means, vars, mcrs)) > > I then take 100 samples of this data frame of varying lengths: > > stats <- lapply (1:100, function (dummy) { > statslist[sample(nrow(statslist), (sample (10:20, 1, replace > = TRUE)), replace = TRUE),]}) > > It returns basically what I want: > > > stats[[i]] > means vars mcrs > 71 1.81 3.832222 2.92725 > 9 2.56 8.127677 4.734874 > 91 3.44 9.66303 5.24902 > 68 0.14 0.1216162 0.008686869 > > except that: > > > is.list(stats[[i]]$means) > [1] TRUE > > I don't want this to be a list, because I want to do > regressions using the variable which are in columns, which > apparently R won't do if they are in lists. > > I have tried every possible combination of working with > do.call, rbind, as.data.frame, etc to get this into matrix > form to no avail. > > I have also tried using sapply instead of lapply, but that > returns vectors which again, R doesn't want to do regression on. > > Any tips would be very much appreciated, been going around in > circles for a while here. > > Edward Waters > PhD Student UNSW > > > -- > View this message in context: > http://n4.nabble.com/Lists-into-matrices-within-lists-again-tp > 1565433p1565433.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >