R help - Oct 2004 - writing lm summary to file?

Hi,

I want to write the summary from a regression.  I am doing this because I do not
see a way of get the std error, tvalues from the coefficients diagnostic. 
n$coef does not give this only get the intercept and slope.  I tried to use
write and write.table and got error in both cases.  I jumped thru the hoops
below to no avail.  Also note, this is in windows.  I used to use unix, and do
not recall this problem, but only have windows at this point.  Any suggestions
are greatly appreciated.

John

> n<-lm(formula = A ~ B, data = y)
> summary(n)
Call:
lm(formula = A ~ B, data = y)

Residuals:
    Min      1Q  Median      3Q     Max 
-5.0165 -0.9726 -0.4707  1.7783  3.8563 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)   
(Intercept)  1.15072    0.38135   3.017  0.00506 **
B        0.18042    0.05254   3.434  0.00171 **
---
Signif. codes:  0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` '
1

Residual standard error: 2.14 on 31 degrees of freedom
Multiple R-Squared: 0.2755,     Adjusted R-squared: 0.2522 
F-statistic: 11.79 on 1 and 31 DF,  p-value: 0.001711 
> help(write.table)
> write.table(summary(n),file ="C\\....")
Error in as.data.frame.default(x[[i]], optional = TRUE) : 
        can't coerce summary.lm into a data.frame
> help(write)
> write(summary(n),file ="C\\..")
Error in cat(list(...), file, sep, fill, labels, append) : 
        argument 1 not yet handled by cat

Visit our website at http://www.ubs.com

This message contains confidential information and is intend...{{dropped}}

On Wed, 27 Oct 2004 John-J.Smith at ubs.com wrote:
> Hi,
> 
> I want to write the summary from a regression.  I am doing this because
> I do not see a way of get the std error, tvalues from the coefficients
> diagnostic.  n$coef does not give this only get the intercept and slope.  
> I tried to use write and write.table and got error in both cases.  I
> jumped thru the hoops below to no avail.  Also note, this is in windows.  
> I used to use unix, and do not recall this problem, but only have
> windows at this point.  Any suggestions are greatly appreciated.
coef(summary(n)) is what you want, I believe.  That's a matrix.
If you want to write the printed output of the summary to a file
(not the same thing at all: see the FAQ Q8.1), use sink().

If you read the help for write() and write.table() I am surprised you 
expected them to work.  Did you try str() on what you were trying to 
write?
> 
> John
> 
> 
> > n<-lm(formula = A ~ B, data = y)
> > summary(n)
> 
> Call:
> lm(formula = A ~ B, data = y)
> 
> Residuals:
>     Min      1Q  Median      3Q     Max 
> -5.0165 -0.9726 -0.4707  1.7783  3.8563 
> 
> Coefficients:
>             Estimate Std. Error t value Pr(>|t|)   
> (Intercept)  1.15072    0.38135   3.017  0.00506 **
> B        0.18042    0.05254   3.434  0.00171 **
> ---
> Signif. codes:  0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 `
' 1
> 
> Residual standard error: 2.14 on 31 degrees of freedom
> Multiple R-Squared: 0.2755,     Adjusted R-squared: 0.2522 
> F-statistic: 11.79 on 1 and 31 DF,  p-value: 0.001711 
> 
> > help(write.table)
> 
> > write.table(summary(n),file ="C\\....")
> Error in as.data.frame.default(x[[i]], optional = TRUE) : 
>         can't coerce summary.lm into a data.frame
> > help(write)
> 
> > write(summary(n),file ="C\\..")
> Error in cat(list(...), file, sep, fill, labels, append) : 
>         argument 1 not yet handled by cat

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595