I have a question about the F test.

In R language, three parameters "two.side", "less" and "greater" can be specified as arguments of var.test (). However, I can not specify a one-sided test like "one.side".

Also, in Microsoft Excel, the F. TEST function was a two-sided test, and the analysis tool's F test was a one-sided test.

I have considered that the one-sided test is mainly used in the F test. Am I wrong?

  • 2
    $\begingroup$ @nekoneko your question is unclear. However in relation to the variance test in R, "less" and "greater" are the two possible one sided alternatives. If you don't specify which one-tailed alternative you're interested in, it isn't one-tailed. Naturally there isn't a "one.sided" without specifying "less" or "greater", and as soon as you do specify which side, it's obviously one-sided. It's exactly the same with a one-sided t-test in R. Take a look at ?t.test. $\endgroup$
    – Glen_b
    May 11 '19 at 2:24
  • $\begingroup$ @glen_b Yeah, that should have been clearer. I deleted it (for now). $\endgroup$ May 12 '19 at 3:31
  • $\begingroup$ Thank you @Glen_b. It seems that I was misunderstood. I do not want to test "greater" and "less". So I have to choose "two.sided". Thank you again. $\endgroup$
    – nekoneko
    May 12 '19 at 20:29
  • $\begingroup$ Glen_b♦ I would like to make your comment the best answer, so can you answer me? $\endgroup$
    – nekoneko
    May 13 '19 at 20:33

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.