How do I extract the standard error of the parameter estimates? Also, if I would like to add two parameters together (x+y), can I use this equation to calculate the new standard error?: x = parameter 1 y = parameter 2 xSE = SE parameter 1 ySE = SE parameter 2 NewSE=(x+y)*sqrt((xSE/x)^2+(ySE/y)^2) -- View this message in context: http://r.789695.n4.nabble.com/How-can-I-extract-information-from-list-which-class-is-nls-tp804151p3476154.html Sent from the R help mailing list archive at Nabble.com.
How can I extract information from list which class is nls
8 messages · Schatzi, Joshua Wiley, Peter Ehlers +2 more
Hi,
One way would be:
summary(nls.object)[["coefficients"]][, "Std. Error"]
If you have a hankering to do it yourself rather than go through the
summary formula, the code here will get you there:
getAnywhere("summary.nls")
If you are going to be doing it a lot, creating a little function might be nice:
se.coef.nls <- function(model) {
summary(model)[["coeffficients"]][, "Std. Error"]
}
se.coef.nls(yourmodel)
I have been doing something along those lines for a number of models.
Sadly, there is not a consistent name always given to the parameter or
coefficients table, so it will not quite be one size fits all, but
generally using str(), you can figure out what the names of what you
want are so it is not much trouble to get out.
str(summary(yourmodel))
Cheers,
Josh
On Tue, Apr 26, 2011 at 11:21 AM, Schatzi <adele_thompson at cargill.com> wrote:
How do I extract the standard error of the parameter estimates? Also, if I would like to add two parameters together (x+y), can I use this equation to calculate the new standard error?: x = parameter 1 y = parameter 2 xSE = SE parameter 1 ySE = SE parameter 2 NewSE=(x+y)*sqrt((xSE/x)^2+(ySE/y)^2) -- View this message in context: http://r.789695.n4.nabble.com/How-can-I-extract-information-from-list-which-class-is-nls-tp804151p3476154.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.
Joshua Wiley Ph.D. Student, Health Psychology University of California, Los Angeles http://www.joshuawiley.com/
On 2011-04-26 15:11, Joshua Wiley wrote:
Hi,
One way would be:
summary(nls.object)[["coefficients"]][, "Std. Error"]
If you have a hankering to do it yourself rather than go through the
summary formula, the code here will get you there:
getAnywhere("summary.nls")
If you are going to be doing it a lot, creating a little function might be nice:
se.coef.nls<- function(model) {
summary(model)[["coeffficients"]][, "Std. Error"]
}
se.coef.nls(yourmodel)
I have been doing something along those lines for a number of models.
Sadly, there is not a consistent name always given to the parameter or
coefficients table, so it will not quite be one size fits all, but
generally using str(), you can figure out what the names of what you
want are so it is not much trouble to get out.
I would generally use the coef() extractor function if it's available (and it is for nls()). ?nls has an example: coef(summary(fm1DNase1)) which is a matrix from which you can get the SEs: coef(summary(fm1DNase1))[,"Std. Error"] or coef(summary(fm1DNase1))[, 2] Schatzi, As to your other question about 'adding two parameters', that doesn't make sense to me. Can you provide a sensible example? In any case you would no doubt have to look at the covariance matrix of the parameter estimates (with vcov()). Peter Ehlers
str(summary(yourmodel)) Cheers, Josh On Tue, Apr 26, 2011 at 11:21 AM, Schatzi<adele_thompson at cargill.com> wrote:
How do I extract the standard error of the parameter estimates? Also, if I would like to add two parameters together (x+y), can I use this equation to calculate the new standard error?: x = parameter 1 y = parameter 2 xSE = SE parameter 1 ySE = SE parameter 2 NewSE=(x+y)*sqrt((xSE/x)^2+(ySE/y)^2) -- View this message in context: http://r.789695.n4.nabble.com/How-can-I-extract-information-from-list-which-class-is-nls-tp804151p3476154.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.
I would generally use the coef() extractor function if it's available (and it is for nls()). ?nls has an example: ?coef(summary(fm1DNase1))
I knew about coef() on model objects, but I did not know it had methods for their summaries. What wonderful information! Josh
which is a matrix from which you can get the SEs: ?coef(summary(fm1DNase1))[,"Std. Error"] or ?coef(summary(fm1DNase1))[, 2] Schatzi, As to your other question about 'adding two parameters', that doesn't make sense to me. Can you provide a sensible example? In any case you would no doubt have to look at the covariance matrix of the parameter estimates (with vcov()). Peter Ehlers
On Tue, Apr 26, 2011 at 2:21 PM, Schatzi <adele_thompson at cargill.com> wrote:
How do I extract the standard error of the parameter estimates? Also, if I would like to add two parameters together (x+y), can I use this equation to calculate the new standard error?: x = parameter 1 y = parameter 2 xSE = SE parameter 1 ySE = SE parameter 2 NewSE=(x+y)*sqrt((xSE/x)^2+(ySE/y)^2)
Try taking the square roots of their variance:
example(nls) sqrt(diag(vcov(fm1DNase1)))
Asym xmid scal 0.07815395 0.08135321 0.03227080
# it gives the same result as another solution # offered on this thread coef(summary(fm1DNase1))[, "Std. Error"]
Asym xmid scal 0.07815395 0.08135321 0.03227080
Statistics & Software Consulting GKX Group, GKX Associates Inc. tel: 1-877-GKX-GROUP email: ggrothendieck at gmail.com
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20110427/ef3a42d5/attachment.pl>
On Apr 27, 2011, at 9:28 AM, Schatzi wrote:
Here is more information on the equation. It is a growth function: Growth = a + b*(1-exp(-k*time)) where a, b and k are parameters. I wanted to test the difference in total growth between treatments and the parameters a + b represent total growth. Thus, I figured that I could add the parameters and the SE to test the treatment differences. I also need the rate constant k and the initial values a and their treatment differences. I found the equation I posted on this website: http://sci.tech-archive.net/Archive/sci.math/2004-07/4780.html This book was referenced: Kendall & Stuart's Advanced Theory of Statistics
You should look at the vignettes and examples in the `fda` package. fda == Functional Data Analysis and the authors have a book that adds further background. http://ego.psych.mcgill.ca/misc/fda/ex-growth-a2.html You might also try: RSiteSearch("growth curve")
David. > > Thanks for all you your help. > Adele > > > From: ml-node+3477116-1659458804-211414 at n4.nabble.com [mailto:ml-node+3477116-1659458804-211414 at n4.nabble.com > ] > Sent: Tuesday, April 26, 2011 09:29 PM > To: Thompson, Adele - Adele_Thompson at cargill.com > Subject: Re: How can I extract information from list which class is > nls > > On Tue, Apr 26, 2011 at 2:21 PM, Schatzi <[hidden email]</user/ > SendEmail.jtp?type=node&node=3477116&i=0&by-user=t>> wrote: > >> How do I extract the standard error of the parameter estimates? >> >> Also, if I would like to add two parameters together (x+y), can I >> use this >> equation to calculate the new standard error?: >> x = parameter 1 >> y = parameter 2 >> xSE = SE parameter 1 >> ySE = SE parameter 2 >> >> NewSE=(x+y)*sqrt((xSE/x)^2+(ySE/y)^2) > > Try taking the square roots of their variance: > >> example(nls) >> sqrt(diag(vcov(fm1DNase1))) > Asym xmid scal > 0.07815395 0.08135321 0.03227080 >> >> # it gives the same result as another solution >> # offered on this thread >> coef(summary(fm1DNase1))[, "Std. Error"] > Asym xmid scal > 0.07815395 0.08135321 0.03227080 > > > -- > Statistics & Software Consulting > GKX Group, GKX Associates Inc. > tel: 1-877-GKX-GROUP > email: ggrothendieck at gmail.com > > ______________________________________________ > [hidden email]</user/SendEmail.jtp?type=node&node=3477116&i=1&by- > user=t> 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. > > ________________________________ > If you reply to this email, your message will be added to the > discussion below: > http://r.789695.n4.nabble.com/How-can-I-extract-information-from-list-which-class-is-nls-tp804151p3477116.html > To unsubscribe from How can I extract information from list which > class is nls, click here<http://r.789695.n4.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=804151&code=YWRlbGVfdGhvbXBzb25AY2FyZ2lsbC5jb218ODA0MTUxfC03MzgxNjM2NTM= > >. > > > ----- > In theory, practice and theory are the same. In practice, they are > not - Albert Einstein > -- > View this message in context: http://r.789695.n4.nabble.com/How-can-I-extract-information-from-list-which-class-is-nls-tp804151p3478123.html > Sent from the R help mailing list archive at Nabble.com. > [[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. David Winsemius, MD West Hartford, CT
An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20110427/6a15b8dc/attachment.pl>