0
$\begingroup$

I have three same scaled variables (Strategy 1: Stay, Strategy 2: Move, Strategy 3: Move far) which I want to compare in a heatmap for different locations. For each strategy an operator earns a specific amount of income (interval scaled). The data is built up like this:

Location   | Strategy 1  | Strategy 2   | Strategy 3 

Location A |      10     |     20       |     30
Location B |      30     |     20       |     10
Location C |      10     |     30       |     20
Location D |     ...     |     ...      |     ...

Now I want to normalize the variables into an ordinal structure to make them comparable in a heatmap, like Strategy 1: -1 < Strategy 2: 0 < Strategy 3: 1

Examples of desired normalized ordinal value:

  • For Location A: 0 < val < 1

  • For Location B: -1 < val < 0

  • For Location C: 0 < val < 1

The ordinal value of one location should be close to the maximum value for one location and be influenced by the income of the second best strategy. The third best strategy should have a low influence.

Is there any practice that allows this, or do you have other suggestions to solve this problem? Thanks!

$\endgroup$
  • $\begingroup$ I did not quite understand the question. What is "row" here? What is ordinal value? Are you thinking of ranking strategies, maybe not? $\endgroup$ – ttnphns Sep 13 '17 at 12:04
  • $\begingroup$ Please register &/or merge your accounts (you can find information on how to do this in the My Account section of our help center), then you will be able to edit & comment on your own question. $\endgroup$ – Silverfish Sep 13 '17 at 13:49
  • $\begingroup$ Hi @ttnphns, I updated the desired output description. I would like to create an ranking of the strategies, but with scalar values. e.g. to create a new kind of "dummy variable" which for instance is -1 for 100% income with the 1st strategy, 0% income with the 2nd strategy and 0% income with the 3rd strategy $\endgroup$ – Sam Sep 13 '17 at 15:45
  • $\begingroup$ Ask to reopen your question (push 'reopen' button). $\endgroup$ – ttnphns Sep 13 '17 at 15:55
  • $\begingroup$ There is no reopen button, just share/cite/edit/delete/flag @peter-flom please reopen the question $\endgroup$ – Sam Sep 13 '17 at 16:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.