2
$\begingroup$

I am currently taking a research methods course online. Today we talked about systematic and random error. The instructor pointed out that systematic error is expected to influence the mean of an estimate, while random error is expected to influence the variance of an estimate, but not the mean.

My observation is that there is a relationship between the sources of error and the effect on the distribution of my estimate:

  • Systematic error affects the 1st moment (the mean)
  • Random error affects the 2nd moment (variance)

Are there other forms of error which can alter other moments, like skew?

$\endgroup$
2
$\begingroup$

Generally, there is an entire probability distribution of random errors, so the random errors also affect skew, kurtosis, and all other (non-location) aspects of the distribution of the measurements. But one must be careful about such "blanket" statements. If you can give more details of what is being measured, how it is being measured etc., the above "blanket" statement could be modified.

$\endgroup$

Your Answer

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

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