[R] values from a linear model

Tue Jul 24 12:40:58 CEST 2007

```

Manuele Pesenti wrote:
>
> Dear R users,
> how can I extrapolate values listed in the summary of an lm model but not
> directly available between object values such as the the standard errors
> of
> the calculated parameters?
>
> for example I got a model:
>
> mod <- lm(Crd ~ 1 + Week, data=data)
>
> and its summary:
>
>> summary(mod)
>
> Call:
> lm(formula = Crd ~ 1 + Week, data = data, model = TRUE, y = TRUE)
>
> Residuals:
>        Min         1Q     Median         3Q        Max
> -4.299e-03 -1.653e-03  2.628e-05  1.291e-03  5.130e-03
>
> Coefficients:
>              Estimate Std. Error  t value Pr(>|t|)
> (Intercept) 1.000e+01  3.962e-04 25238.73   <2e-16 ***
> Week        5.038e-04  6.812e-06    73.96   <2e-16 ***
> ---
> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
>
> Residual standard error: 0.001966 on 98 degrees of freedom
> Multiple R-Squared: 0.9824,     Adjusted R-squared: 0.9822
> F-statistic:  5469 on 1 and 98 DF,  p-value: < 2.2e-16
>
> I'm interested in values of Std. Error of coefficients...
>
> thank you very much
>

If you want to assign these values to some other variables, try assigning
the result of the summary() to a variable and working with its components
(the result is a list, use \$ or [[]] to get its members)

mod.sum<-summary(mod)
then
coef(mod.sum)[,2]
or
mod.sum\$coefficients[,2]

will give you those Std. Errors

--
View this message in context: http://www.nabble.com/values-from-a-linear-model-tf4134911.html#a11760459
Sent from the R help mailing list archive at Nabble.com.

```