An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120228/d0dedf09/attachment.pl>
Quantile scores as dependent variables.. an R and general method question
5 messages · Rob James, ilai, Doran, Harold
On Tue, Feb 28, 2012 at 3:54 PM, Rob James <aetiologic at gmail.com> wrote:
I have a dataset that does not include native scores, but only serial quantile rankings for a set of units. Clearly these observations are dependent (in that you can't alter one observation without also altering others). Are there methods for dealing with quantile dependent variables. My atempt to find such methods has not bee successful.
Really? because google found 227k hits for "R quantile regression" - none of them lead anywhere ?
Any leads to theory, texts or R code would be most appeciated. ? ? ? ?[[alternative HTML version deleted]]
______________________________________________ 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.
Rob The OP is looking for a way to deal with outcomes scores that are quantiles, not a method that models different quantiles of the conditional distribution where an outcome is a continuous variable. So, I don't think QR methods is what is needed.
From: r-help-bounces at r-project.org [r-help-bounces at r-project.org] On Behalf Of ilai [keren at math.montana.edu]
Sent: Wednesday, February 29, 2012 1:30 PM
To: Rob James
Cc: r-help at r-project.org
Subject: Re: [R] Quantile scores as dependent variables.. an R and general method question
Sent: Wednesday, February 29, 2012 1:30 PM
To: Rob James
Cc: r-help at r-project.org
Subject: Re: [R] Quantile scores as dependent variables.. an R and general method question
On Tue, Feb 28, 2012 at 3:54 PM, Rob James <aetiologic at gmail.com> wrote: > I have a dataset that does not include native scores, but only serial > quantile rankings for a set of units. > > Clearly these observations are dependent (in that you can't alter one > observation without also altering others). > > Are there methods for dealing with quantile dependent variables. My atempt > to find such methods has not bee successful. > Really? because google found 227k hits for "R quantile regression" - none of them lead anywhere ? > Any leads to theory, texts or R code would be most appeciated. > > [[alternative HTML version deleted]] > > ______________________________________________ > 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. ______________________________________________ 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.
On Wed, Feb 29, 2012 at 1:23 PM, Doran, Harold <HDoran at air.org> wrote:
The OP is looking for a way to deal with outcomes scores that are quantiles, not a method that models different quantiles of the conditional distribution where an outcome is a continuous variable. So, I don't think QR methods is what is needed.
Huhh? First, "deal with" is rather vague, does it mean summary? plot? delete? as the OP is worried about independence, some modelling exercise is more likely. Second, where do you think the OP quantiles come from if not from the conditional distribution of some continuous RV ? hint: not France. So for "...leads to theory, texts or R code.." on the distributional properties of quantiles/order statistics, do you have a better suggestion for a starting point than QR methods? Cheers
________________________________________ From: r-help-bounces at r-project.org [r-help-bounces at r-project.org] On Behalf Of ilai [keren at math.montana.edu] Sent: Wednesday, February 29, 2012 1:30 PM To: Rob James Cc: r-help at r-project.org Subject: Re: [R] Quantile scores as dependent variables.. an R and general method question On Tue, Feb 28, 2012 at 3:54 PM, Rob James <aetiologic at gmail.com> wrote: I have a dataset that does not include native scores, but only serial quantile rankings for a set of units. Clearly these observations are dependent (in that you can't alter one observation without also altering others). Are there methods for dealing with quantile dependent variables. My atempt to find such methods has not bee successful. Really? because google found 227k hits for "R quantile regression" - none of them lead anywhere ? Any leads to theory, texts or R code would be most appeciated. ? ? ? ?[[alternative HTML version deleted]] ______________________________________________ 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. ______________________________________________ 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.
Typically this list doesn't support general statistical questions and unfortunately I don't have a better recommendation. It may be more helpful for you to work with a statistician than seek help here. My point is simply that quantile regression is not for modeling outcomes that are quantiles. The fact that you have a dependent variables expressed as a percentile doesn't mean that quantile regression is the appropriate approach.
From: ilaik9 at gmail.com [ilaik9 at gmail.com] On Behalf Of ilai [keren at math.montana.edu]
Sent: Wednesday, February 29, 2012 5:52 PM
To: Doran, Harold
Cc: Rob James; r-help at r-project.org
Subject: Re: [R] Quantile scores as dependent variables.. an R and general method question
Sent: Wednesday, February 29, 2012 5:52 PM
To: Doran, Harold
Cc: Rob James; r-help at r-project.org
Subject: Re: [R] Quantile scores as dependent variables.. an R and general method question
On Wed, Feb 29, 2012 at 1:23 PM, Doran, Harold <HDoran at air.org> wrote: > > The OP is looking for a way to deal with outcomes scores that are quantiles, not a method that models different quantiles of the conditional distribution where an outcome is a continuous variable. So, I don't think QR methods is what is needed. Huhh? First, "deal with" is rather vague, does it mean summary? plot? delete? as the OP is worried about independence, some modelling exercise is more likely. Second, where do you think the OP quantiles come from if not from the conditional distribution of some continuous RV ? hint: not France. So for "...leads to theory, texts or R code.." on the distributional properties of quantiles/order statistics, do you have a better suggestion for a starting point than QR methods? Cheers > ________________________________________ > From: r-help-bounces at r-project.org [r-help-bounces at r-project.org] On Behalf Of ilai [keren at math.montana.edu] > Sent: Wednesday, February 29, 2012 1:30 PM > To: Rob James > Cc: r-help at r-project.org > Subject: Re: [R] Quantile scores as dependent variables.. an R and general method question > > On Tue, Feb 28, 2012 at 3:54 PM, Rob James <aetiologic at gmail.com> wrote: >> I have a dataset that does not include native scores, but only serial >> quantile rankings for a set of units. >> >> Clearly these observations are dependent (in that you can't alter one >> observation without also altering others). >> >> Are there methods for dealing with quantile dependent variables. My atempt >> to find such methods has not bee successful. >> > > Really? because google found 227k hits for "R quantile regression" - > none of them lead anywhere ? > > >> Any leads to theory, texts or R code would be most appeciated. >> >> [[alternative HTML version deleted]] >> >> ______________________________________________ >> 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. > > ______________________________________________ > 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.