Say I have two sample distributions where for each distribution I know:

  • Number of Values: $n_1, n_2$.
  • The means: $m_1, m_2$.
  • Standard Deviation: $s_1, s_2$.

With this information I am able to calculate a $t$-value, $t_s$, using Welch's $t$-test (we assume the two distribution do not share the same variance).

If we were to add a single number to each sample distribution and recalculate the t-value, what would be the most efficient way to do it? I know based off: How to add and subtract values from an average? and incremental computation of standard deviation that it is possible to incrementally update the mean and standard deviation. I know I could plug in the formulas from each page to find a solution, however I was looking for something in terms of $t_s$. I am new to stats and I am wondering if there is a name for all of this that I can look up.


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.