0
$\begingroup$

I have a panel data set - a score on a single measure for hundreds of people, taken once per day for all 365 days of a year. I'm looking for the best method to assess overall variation within each person's scores. In other words, I care about the overall frequency and magnitude of changes, not just the absolute change between Day 1 and Day 365.

It seems like simply the variance (or SD) for each individual would be a fairly decent indicator, but this doesn't seem to be used in panel data studies. Is there something better I should use?

I'll then use this (unknown) indicator as a DV in a regression, to see how different personal characteristics influence the degree of variation throughout the year.

Thanks!

Improve this question
$\endgroup$
2
  • $\begingroup$ I too think variance or SD should serve your need. why do you think variance or SD doesn't seem to be used in panel studies? $\endgroup$
    – rsl
    Commented Jun 3, 2016 at 6:52
  • $\begingroup$ I'm just getting into panel data research and hadn't yet seen a study using SD. $\endgroup$
    – DaGu
    Commented Jul 19, 2016 at 18:34

1 Answer 1

Reset to default
0
$\begingroup$

You could consider the coefficient of variation which is defined as the ratio of the standard deviation to the mean. Reference: https://en.wikipedia.org/wiki/Coefficient_of_variation

Improve this answer
$\endgroup$
0

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.