I am taking samples from a large population. The samples can be of five, ten, or twenty items.

I have read that Tukey's trimean is better than the median. Is that true? Are there any other estimators better than the median which are simple enough to calculate on a spreadsheet? I read that the median is the best single-point estimator, but the three-point trimean is better. On the other hand, I imagine that calculating the trimean from a sample size of five cannot be reliable.

If it helps, the samples are taken from sequential items for quality control, so the "average" of one sample is likely to be autocorrelated with the "average" in the next sample. The trade-off is that smaller samples lead to a quicker response but with more false alarms than bigger samples.


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.