Parameter info from nls object

[Dobrozemsky Georg]

Hi!

When checking validity of a model for a large number
of experimental data I thought it to be interesting
to check the information provided by
the summary method programmatically.

Still I could not find out which method to
use to get to those data.

Example (not my real world data, but to show the point):

[BEGIN]
> DNase1 <- subset(DNase, Run == 1)
> fm1DNase1 <- nls(density ~ SSlogis(log(conc), Asym, xmid, scal), DNase1)
> summary(fm1DNase1)

Formula: density ~ SSlogis(log(conc), Asym, xmid, scal)

Parameters:
     Estimate Std. Error t value Pr(>|t|)    
Asym  2.34518    0.07815   30.01 2.17e-13 ***
xmid  1.48309    0.08135   18.23 1.22e-10 ***
scal  1.04146    0.03227   32.27 8.51e-14 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.01919 on 13 degrees of freedom

Number of iterations to convergence: 0
Achieved convergence tolerance: 3.302e-06
[END]

now i would like something like
> property(fm1DNase1,'Asym','StErr')
[1] 0.07815

> property(fm1DNase1,'xmid','Pr')
[1] 1.22e-10

I hope you understand the point (and yes,
I read manuals, tried help(), read through
the archives), sorry if still this is trivial :P

wfG

George

+---
| Dr. Georg Dobrozemsky
| Dept. of. Nuclear Medicine
| Med. Univ. Vienna
+---


        [[alternative HTML version deleted]]


______________________________________________
R-help@... 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]

On Nov 9, 2009, at 12:30 PM, Dobrozemsky Georg wrote:

> Hi!
>
> When checking validity of a model for a large number
> of experimental data I thought it to be interesting
> to check the information provided by
> the summary method programmatically.
>
> Still I could not find out which method to
> use to get to those data.
>
> Example (not my real world data, but to show the point):
>
> [BEGIN]
>> DNase1 <- subset(DNase, Run == 1)
>> fm1DNase1 <- nls(density ~ SSlogis(log(conc), Asym, xmid, scal),  
>> DNase1)
>> summary(fm1DNase1)
>
> Formula: density ~ SSlogis(log(conc), Asym, xmid, scal)
>
> Parameters:
>     Estimate Std. Error t value Pr(>|t|)
> Asym  2.34518    0.07815   30.01 2.17e-13 ***
> xmid  1.48309    0.08135   18.23 1.22e-10 ***
> scal  1.04146    0.03227   32.27 8.51e-14 ***
> ---
> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
>
> Residual standard error: 0.01919 on 13 degrees of freedom
>
> Number of iterations to convergence: 0
> Achieved convergence tolerance: 3.302e-06
> [END]
>
> now i would like something like
>> property(fm1DNase1,'Asym','StErr')
> [1] 0.07815
>
>> property(fm1DNase1,'xmid','Pr')
> [1] 1.22e-10

Applying str to the summary object should have given you sufficient  
clues. Then it is a simple matter of extracting the right component,  
and then properly extracting from the matrix with the correct names:

 > str(summary(fm1DNase1)$parameters)
  num [1:3, 1:4] 2.3452 1.4831 1.0415 0.0782 0.0814 ...
  - attr(*, "dimnames")=List of 2
   ..$ : chr [1:3] "Asym" "xmid" "scal"
   ..$ : chr [1:4] "Estimate" "Std. Error" "t value" "Pr(>|t|)"

 > summary(fm1DNase1)$parameters['Asym',"Std. Error"]
[1] 0.07815395
 > summary(fm1DNase1)$parameters['xmid',"Pr(>|t|)"]
[1] 1.218535e-10

>
> I hope you understand the point (and yes,
> I read manuals, tried help(), read through
> the archives), sorry if still this is trivial :P
>
> wfG
>
> George
>
> +---
> | Dr. Georg Dobrozemsky
> | Dept. of. Nuclear Medicine
> | Med. Univ. Vienna
> +---
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@... 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
Heritage Laboratories
West Hartford, CT

______________________________________________
R-help@... 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]

On Nov 9, 2009, at 12:59 PM, David Winsemius wrote:

>
> On Nov 9, 2009, at 12:30 PM, Dobrozemsky Georg wrote:
>
>> Hi!
>>
>> When checking validity of a model for a large number
>> of experimental data I thought it to be interesting
>> to check the information provided by
>> the summary method programmatically.
>>
>> Still I could not find out which method to
>> use to get to those data.
>>
>> Example (not my real world data, but to show the point):
>>
>> [BEGIN]
>>> DNase1 <- subset(DNase, Run == 1)
>>> fm1DNase1 <- nls(density ~ SSlogis(log(conc), Asym, xmid, scal),  
>>> DNase1)
>>> summary(fm1DNase1)
>>
>> Formula: density ~ SSlogis(log(conc), Asym, xmid, scal)
>>
>> Parameters:
>>    Estimate Std. Error t value Pr(>|t|)
>> Asym  2.34518    0.07815   30.01 2.17e-13 ***
>> xmid  1.48309    0.08135   18.23 1.22e-10 ***
>> scal  1.04146    0.03227   32.27 8.51e-14 ***
>> ---
>> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
>>
>> Residual standard error: 0.01919 on 13 degrees of freedom
>>
>> Number of iterations to convergence: 0
>> Achieved convergence tolerance: 3.302e-06
>> [END]
>>
>> now i would like something like
>>> property(fm1DNase1,'Asym','StErr')
>> [1] 0.07815
>>
>>> property(fm1DNase1,'xmid','Pr')
>> [1] 1.22e-10
>
> Applying str to the summary object should have given you sufficient  
> clues. Then it is a simple matter of extracting the right component,  
> and then properly extracting from the matrix with the correct names:

I should probably admit that I have gotten into bad habits over the  
couple of years I have used R, and that the "approved" method is to  
use the extractor function, coef (which is one of the links from the  
page regarding either nls or summary.nls (but not from just the  
summary help page.)

 > coef(summary(fm1DNase1))
      Estimate Std. Error  t value     Pr(>|t|)
Asym 2.345180 0.07815395 30.00719 2.165502e-13
xmid 1.483090 0.08135321 18.23026 1.218535e-10
scal 1.041455 0.03227080 32.27236 8.506915e-14

 > coef(fm1DNase1)
     Asym     xmid     scal
2.345180 1.483090 1.041455

>
> > str(summary(fm1DNase1)$parameters)
> num [1:3, 1:4] 2.3452 1.4831 1.0415 0.0782 0.0814 ...
> - attr(*, "dimnames")=List of 2
>  ..$ : chr [1:3] "Asym" "xmid" "scal"
>  ..$ : chr [1:4] "Estimate" "Std. Error" "t value" "Pr(>|t|)"
>
> > summary(fm1DNase1)$parameters['Asym',"Std. Error"]
> [1] 0.07815395
> > summary(fm1DNase1)$parameters['xmid',"Pr(>|t|)"]
> [1] 1.218535e-10
>
>
>>
>> I hope you understand the point (and yes,
>> I read manuals, tried help(), read through
>> the archives), sorry if still this is trivial :P
>>
>> wfG
>>
>> George

--
David Winsemius, MD
Heritage Laboratories
West Hartford, CT

______________________________________________
R-help@... 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.