0
$\begingroup$

I have played around with logistic regression a little using movement data intervals that are prelabeled as either resting or active. I now found that if I divide the mean movement of the individual intervals by the intervals standard deviation, the outcome is quite a good predictor of whether the interval is a resting interval or or active, with an average auc = 0.93 in a 20 fold cross validation.

Does someone have an idea of what I have created dividing the mean by the standard deviation? Its like a flipped coefficient of variation, or semi-normalization?!.

I want to report this in an essay, so I am asking myself whether there is a name to this statistic.

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

You have created what can be called a normalised mean. Normalised in the sense that it corresponds to a related random variable which now has $\sigma = 1$ (since you divide by standard deviation, look it up in any statistics textbook).

The normalised mean can be a better predictor resulting from its normalisation which makes it robust to the range of fluctuations.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thank you so much for this. That makes a lot of sense. $\endgroup$
    – DataVader
    Commented May 26, 2021 at 15:00

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.