R help - Jul 2003 - P-value for F from summary.lm (was RE: (no subject))

[Please use the subject line!]

In the help page for summary.lm, the "Value" section says that the
returned
object has a component called "fstatistic", which has the F-statistic
and
the associated numerator and denominator degrees of freedom.  You can get
the p-value by something like:

  fstat <- summary(speciallinearmodel)$fstatistic
  pval <- pf(fstat[1], fstat[2], fstat[3], lower.tail=FALSE)

HTH,
Andy
> -----Original Message-----
> From: joerg-burmester at gmx.net [mailto:joerg-burmester at gmx.net] 
> Sent: Monday, July 07, 2003 7:50 AM
> To: R-help at stat.math.ethz.ch
> Cc: amaitour at pasteur.fr; bitwrit at ozemail.com.au; 
> p.dalgaard at biostat.ku.dk; gregor.gawron at rmf.ch; 
> mditzen at zedat.fu-berlin.de
> Subject: [R] (no subject)
> 
> 
> hi,
> 
> can anyone help me about the simple data probability-value of 
> the F-test, respectively p-value, shown in the summary of a 
> linear model? somehow i can not retrieve this data out of R+ 
> into my sql-dataset. i can get all other data shown with the command: 
> summary(speciallinearmodel) - will have a gui screen
> 
> Call:
> lm(formula = y21 ~ x4 + x18 + x4:x18, data = 
> clustertransformationstabelle)
> Residuals:
> Min       1Q   Median       3Q      Max 
> -0.20915 -0.07354 -0.01823  0.04414  0.45191 
> Coefficients:
> Estimate Std. Error t value Pr(>|t|)    
> (Intercept) -0.003613   0.043142  -0.084 0.933450    
> x4           0.142373   0.113036   1.260 0.211057    
> x18          0.250261   0.127277   1.966 0.052314 .  
> x4:x18       0.818618   0.218541   3.746 0.000315 ***
> ---
> Signif. codes:  0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 `
' 1
> Residual standard error: 0.1305 on 91 degrees of freedom
> Multiple R-Squared: 0.7865,     Adjusted R-squared: 0.7795 
> F-statistic: 111.7 on 3 and 91 DF,  p-value: < 2.2e-16 
> 
> but the desired p-value like in this example p-value: < 
> 2.2e-16, i somehow can not retrieve by any command like 
> attributes(speciallinearmodel). has anyone an idea to get 
> this shown but not selectable respectively exportable p-value?
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list 
> https://www.stat.math.ethz.ch/mailman/listinfo> /r-help
> 
------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments, ...{{dropped}}

"Liaw, Andy" <andy_liaw at merck.com> writes:
> [Please use the subject line!]
> 
> In the help page for summary.lm, the "Value" section says that
the returned
> object has a component called "fstatistic", which has the
F-statistic and
> the associated numerator and denominator degrees of freedom.  You can get
> the p-value by something like:
> 
>   fstat <- summary(speciallinearmodel)$fstatistic
>   pval <- pf(fstat[1], fstat[2], fstat[3], lower.tail=FALSE)
Also, note that this gets printed using format.pval(), which avoids
misleading printouts if the p value underflows to zero:
> format.pval(0)
[1] "< 2.22e-16"

> > but the desired p-value like in this example p-value: < 
> > 2.2e-16, i somehow can not retrieve by any command like 
> > attributes(speciallinearmodel). has anyone an idea to get 
> > this shown but not selectable respectively exportable p-value?

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)             FAX: (+45) 35327907