1

So I have a test that you only get right if you guess correctly three independent variables. Guess any of them wrong or even any combination of the variables wrong and you loose. My question is what model do I try to make sense of this and how do I try to find a correlation between the independent variables?

Context: This is a game I'm trying to model, I have three sensory inputs signaled and a player has to count how many times each of them appeared and he wins if he counted them all correctly.

Paulo Costa
  • 13

1 Answers1

1

If you have only the binary test result for "all three correct", it will be difficult to obtain much information about the success correlation between two variables, because this binary test result doesn't tell you which tests the player got wrong.

So I presume that you have also access to the results of each of the three tests. And then it is just the computation of ordinary correlation coefficients between all three possible pairs of tests. But that would not use the game result (whether all three are correct at the same time).

frank
  • 10,797