Closed vspinu closed 6 years ago
I don't know what you're talking about. At no point is a y~y
plot generated in, say, this output:
fit <- lm(Ozone ~ Solar.R + Wind + Temp, data=airquality)
visreg(fit)
Can you give me an example where you see the plot you're talking about?
I see. It has to do with variable names. Here you go:
tt <- airquality
tt[["Ozone1"]] <- 1:nrow(tt)
fit <- lm(Ozone ~ Solar.R + scale(Wind) + Temp + Ozone1, data=tt)
visreg(fit)
Probably a regexp or match
issue.
Ah ha, very interesting. Yes, it's a grep
issue. Thanks for pointing this out!
Thanks!
Currently the first plot generated by
visreg(fit)
is the response vs response (y~y
) plot which is always tilted. To my mind this plot is not particularly useful. All displays incar
andeffects
packages don't ploty~y
but stick with the actual (right side) terms. I think this plot could be dropped or, maybe, replaced with fitted ~ response plot.