Re: Help with a T-Test involving uncertainties

Ramesh...

If you are going to insist on going this route... ignoring my
suggestion... then at least do the math correctly.

Hint... think about your Step 2 again.

Hint... think hard about the degrees of freedom. The answer is not
four.

OMU



Ramesh Rebba wrote:
1) Take delta, the difference between the sample means of each set (Set1-A,B; Set2-C,D,E).

2) pooled variance will be s1^2/2+s2^2/3 + 2*13% where s1 and s2 are sample standard deviations of each set.

3) T-statistic is (delta)/sqrt(pooled variance) with 2+3-1 = 4 degrees of freedom.

Note that 2*13% represents twice the uncertainty level that you mentioned.

-Ramesh Rebba

.

Relevant Pages

  • Re: Cliki.net a list of packages? Was: Reddit Guys on the Pros and Cons of Lisp
    ... >>>If anybody actually wants to listen to newcomers, rather than ignoring ... >>>any hint of deficiency in their favorite language, ...
    (comp.lang.lisp)
  • Re: Using Exchange as a smarthost
    ... thanks to Bharat for his good suggestion. ... Here I am providing you a hint on Steps 3. ... When you are creating the SMTP connector, ...
    (microsoft.public.exchange.admin)
  • Re: Heidfeld
    ... Of course he's ignoring all the evidence in testing and on track and ... Is that sarcasm, ... Another hint: 2005 and 2006 Renault. ...
    (rec.autos.sport.f1)
  • Re: Blocking low ?
    ... Ignoring the obvious could set the standards of ... meteorology back by 130 yea... ... hint as to whether your method is even vaguely working or not. ...
    (uk.sci.weather)
  • Re: Index View
    ... Thanks for your Suggestion. ... Please Give me the Hint How Can I Proceed? ... Also check how this query affects ... > I expect that you'll find that materializing the view is not the magic ...
    (microsoft.public.sqlserver.programming)