Rationale for df in glance

AndyZ · May 29, 2020, 8:47pm

Hi

I was wondering about the rationale for the df values outputted by the glance() function for outputting an lm object. I understand that what is being outputted are the number of model coefficients being estimated (call this k), but why not k-1 since glance() is also outputting information relevant to the F-test?

Thanks
Andy

mara · June 3, 2020, 3:36pm

There's an open issue/discussion in the broom repo related to this you might take a look at:

github.com/tidymodels/broom

Degrees of freedom is incorrect in the glance function

opened 08:25PM - 26 Jan 18 UTC

closed 09:06PM - 15 Aug 20 UTC

jbryer

consistency/specification

Note that the `summary` function correctly shows 2 and 29 degrees of freedom for… the F, but the `glance` function returns 3 and 29. ``` > library(broom) > data(mtcars) > lm.out <- lm(mpg ~ disp + hp, data = mtcars) > summary(lm.out) Call: lm(formula = mpg ~ disp + hp, data = mtcars) Residuals: Min 1Q Median 3Q Max -4.7945 -2.3036 -0.8246 1.8582 6.9363 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 30.735904 1.331566 23.083 < 2e-16 *** disp -0.030346 0.007405 -4.098 0.000306 *** hp -0.024840 0.013385 -1.856 0.073679 . --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 3.127 on 29 degrees of freedom Multiple R-squared: 0.7482, Adjusted R-squared: 0.7309 F-statistic: 43.09 on 2 and 29 DF, p-value: 2.062e-09 > glance(lm.out) r.squared adj.r.squared sigma statistic p.value df logLik AIC BIC deviance 1 0.7482402 0.7308774 3.126601 43.09458 2.062068e-09 3 -80.30928 168.6186 174.4815 283.4934 df.residual 1 29 ```

AndyZ · June 3, 2020, 4:06pm

Thanks Mara and RStudio

Even after googling I did not find this issue. Either my brain, or googling ability has diminished. Sounds like it will be addressed in the next broom update and is already addressed in the dev version.

Thanks again!
Andy

system · June 24, 2020, 4:06pm

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.