Page 1 of 1

One tail t-statistics

Posted: Thu May 14, 2009 6:26 am
by financier
Hi everyone
I am working on my thesis and I am running regression analysis. I am trying to do one tail t-test but I am only getting results for two tail test. Can anyone tell is there any command or function that I can use to do only one tail t-test. I will really appreciate if someone can help me.

Regards
Financier

Re: One tail t-statistics

Posted: Thu May 14, 2009 6:28 am
by startz
Hi everyone
I am working on my thesis and I am running regression analysis. I am trying to do one tail t-test but I am only getting results for two tail test. Can anyone tell is there any command or function that I can use to do only one tail t-test. I will really appreciate if someone can help me.

Regards
Financier
Just look up the one-tailed critical value in a t-table.

Re: One tail t-statistics

Posted: Thu May 14, 2009 10:37 am
by EViews Glenn
I'll add to Startz's comment by noting that you can use the @qtdist function to get the appropriate critical value (without having to look things up in a table). For example, the (upper) one-tailed .05 critical value may be obtained by @qtdist(.95, df), where df is the degrees-of-freedom. I'll also point out that if EViews is reporting a p-value for a two-tailed test and you wish to know the corresponding one-tailed value, you can always divide by 2.