4
$\begingroup$

I'm making a comparison between the model I build and the data (empirical distribution), using various plots, like this:

enter image description here

  1. Histogram and probability densiti function (PDF)

  2. CDF

  3. CDF at different axis (Weibull Axis)

The bule color is the data (empirical distribution), the black and red curve is the model.

The problem is that, what color scheme should I chose?

I mean, in the 1st plot, I use black to represent model, but in 2nd, and 3rd plot, I use red instead, because it contrasts better with blue dots. I'm afraid this inconsistancy would result in confusion, because there are only 2 type of data (Model vs. Empirical), yet I used 3 colors:

UPDATE:

After changing the black to red

enter image description here

Improve this question
$\endgroup$
3
  • 2
    $\begingroup$ I agree with your consistency argument, and would probably change the black curve in the leftmost plot to red (I think changing the red to black in the other plots would make them harder to read). Otherwise I think these look pretty good. $\endgroup$
    – Matthew Drury
    Jun 4, 2016 at 2:28
  • $\begingroup$ @MatthewDrury Yes, I tried changing the red to black. It's just hard to read. I'll update with your suggestion. $\endgroup$
    – ZK Zhao
    Jun 4, 2016 at 3:01
  • 2
    $\begingroup$ If there is only the slightest chance of your results getting printed, I would ensure the colors also work when the print is in grayscale. In such cases, a very, very light gray for the data distribution, a thicker line in medium gray for the model, and black dots (or other symbols) usually do the trick for me. $\endgroup$
    – geekoverdose
    Jun 14, 2016 at 11:00

1 Answer 1

Reset to default
3
$\begingroup$

I agree with your consistency argument, and would probably change the black curve in the leftmost plot to red (I think changing the red to black in the other plots would make them harder to read). Otherwise I think these look pretty good.

Improve this answer
$\endgroup$
1
  • 2
    $\begingroup$ I've copied this comment by @MatthewDrury as a community wiki answer because the comment is, more or less, an answer to this question. We have a dramatic gap between answers and questions. At least part of the problem is that some questions are answered in comments: if comments which answered the question were answers instead, we would have fewer unanswered questions. $\endgroup$
    – mkt
    Jun 21, 2019 at 16:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.