3
$\begingroup$

I need to calculate the difference between six time series in three ways:

  • Time Series are: Kamel, Dumper, Graben, Traktor, Generator
  • Methods are: Euclidean distance, Manhattan distance and maximum distance

For example, I show the differences between these time series for Euclidean distance in a table like this:

enter image description here

(I have two more such tables.) Now I want to visualize these tables in one diagram, with a different color for each method. What data visualization method should I use?

$\endgroup$
2
  • 1
    $\begingroup$ Who is the audience? What is the key message that you want the audience to get from the graphic? Why are the numbers, their magnitude and significant digits the right ones to use? $\endgroup$ Commented Jul 13, 2014 at 14:51
  • $\begingroup$ The radar plot is good for comparing equivalent metrics on different classes. You could have six lines on 3 radar plots. It might be more informative. $\endgroup$ Commented Jul 13, 2014 at 15:00

2 Answers 2

8
$\begingroup$

The most natural representation would be three heat maps with the same coloring (small multiples).

enter image description here

If the distance scales are too different for a single color scale, you can either use different colors for each or standardize the measures to put them on the same scale.

If there is some consistent ranking of pairs, then overlaid line plots or dot plots may help show the pattern. However, it's no good in these examples with random data.

enter image description here enter image description here

$\endgroup$
1
$\begingroup$

If you want to color code the values in a nice color scale you can try a heatmap plot (see this link). If you are still interested in different colors for each method you can try grouped bars; that will allow you to specify separate colors for each bar (see link).

$\endgroup$

Your Answer

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

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