I know that there are formulas for calculating variance in dynamic fashions. I have also seen formulas for calculating rolling variances assuming you know what number is leaving and entering the window. Is there a formula for calculating variance after removing one or more observations? I am assuming I won't have access to the full dataset, just the observation I am removing and a handful of aggregated features of the dataset.

  • $\begingroup$ What is your question? $\endgroup$ – deemel Apr 22 '19 at 20:00
  • $\begingroup$ The formula is algebraically equivalent to updating the variance when adjoining a single value to a dataset. $\endgroup$ – whuber Apr 23 '19 at 2:09

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.