What is the estimated variance of residuals? From R

Question

I am trying to figure out what is the estimated variance (i.e. the estimated "error") of residuals around a fitted line.

> summary(model)

Call:
lm(formula = fecundity ~ Organic)

Residuals:
    Min      1Q  Median      3Q     Max 
-2.2909 -1.6439 -0.4606  1.5121  3.7273 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  47.6667     1.4907   31.98 9.97e-10 ***
Organic      -8.6788     0.4805  -18.06 9.06e-08 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 2.182 on 8 degrees of freedom
Multiple R-squared:  0.9761,    Adjusted R-squared:  0.9731 
F-statistic: 326.2 on 1 and 8 DF,  p-value: 9.063e-08

> anova(model)
Analysis of Variance Table

Response: fecundity
          Df Sum Sq Mean Sq F value    Pr(>F)    
Organic    1 1553.5 1553.50  326.22 9.063e-08 ***
Residuals  8   38.1    4.76                      
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Did you see this line in the output "Residual standard error: 2.182 on 8 degrees of freedom"? — Aksakal, Feb 09 '15 at 20:58
The variance of the residuals would be $\frac{\text{Sum Sq}}{\text{df}}=\frac{38.1}{8}=4.7625$, or `summary(model)$sigma^2`. — Antoni Parellada, Jan 18 '17 at 17:35

score 4 · Answer 1 · answered Feb 09 '15 at 21:02

4

Did you see this line in the output "Residual standard error: 2.182 on 8 degrees of freedom"?

There's also a line "Residuals" in ANOVA output with "Mean Sq" column.

answered Feb 09 '15 at 21:02

Aksakal

55,939
5
90
176

What is the estimated variance of residuals? From R

1 Answers1