love your visualizations @krstoffr - but this http://rpsychologist.com/d3/correlation/ implies a bit that corr relates to slope of line of best fit...?
-
any chance you could implement moving the fitted line & data points for one corr coeff, too? :)
I can make one later today that shows this. See also my collection here: http://emilkirkegaard.dk/understanding_statistics/ … All made in Shiny with public code. :)
-
-
that would be very neat! and thanks - will definitely check them out!
-
I usually show this manually by dragging a point so var(y) increases, correlation decrease, while slope is the same.
-
but, yeah, I could implement it more clearly if that's a common use case :)
-
Your viz would be better than mine (not easy to do dragable points in Shiny). If you could just add regressions stats, would be ideal.
-
Thinking of: r, slope, intercept, mean prediction error, mean/median squared/absolute residual and so on.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.