add method for summary.lm ($coef)

[bquast]

There is still a slight problem in that it prints

Pr(>|t|)

as:

Pr(¿—t—)

in LaTeX documents

[bquast]

library(printr)
lm(speed ~ dist, data=cars)

            Estimate   Std. Error    t value   Pr(>|t|)

---

(Intercept) 8.2839056 0.8743845 9.473985 0
dist 0.1655676 0.0174945 9.463990 0

Note that it's GFM that doesn't render this table well, it works fine in R

[yihui]

For the LaTeX problem, did you have \usepackage[T1]{fontenc} in the preamble?

[yihui]

I'd suggest a different form of printing, e.g. print out the regression equation, instead of the identical output as the summary.lm method.

[bquast]

That is a good idea, I wasn't sure what to do with the summary one but that makes sense. I'll update this now.

[bquast]

Let me know what you think about this.

[yihui]

I mean lm gives the equation of the form dist = ? + ? speed (instead of the R call), and summary.lm gives the coefficient matrix.

[bquast]

I'm using LyX which I'm quite sure does this when rendering using pdflatex, it also occurs when using LuaTeX.

This may be related, when I produce a kable in an ERT in standard text it always works fine, but the if the first number is negative (i.e. coefficient of the intercept) and I embed it in a table float, in fails to render.