Discussion about math, puzzles, games and fun. Useful symbols: ÷ × ½ √ ∞ ≠ ≤ ≥ ≈ ⇒ ± ∈ Δ θ ∴ ∑ ∫ π -¹ ² ³ °

You are not logged in.

- Topics: Active | Unanswered

Pages: **1**

**M Al Hor****Guest**

Hi all,

I want to ask about the relationship between p-value and confidence intervals (CIs). Suppose that I have two 95% confidence intervals and I want to see if they are statistically significant or not. How can I do that? Is the fact that they do not overlap means that p-value <0.05? what about if they do overlap?

Thanks in advance

Muhammad Al Hor

**liuv****Member**- Registered: 2006-05-14
- Posts: 29

1-5%*5%=99.75%

I'm from Beijing China.

Offline

**M Al Hor****Guest**

Sorry but I did not get. You can please clarify your answer.

Thanks

**Mathaphobe****Member**- Registered: 2006-05-20
- Posts: 18

I think he meant 1munis 5percent multiply by 5percent equals 99.75 percent

if im stateing the odvious please tell me..

And what is a p-value? or a Cls?

Do you like my new avatar? (even if you don't i do )

(it's victor from oh my gods)

-An Incremental Development

Offline

**George,Y****Member**- Registered: 2006-03-12
- Posts: 1,306

statistically significant means initially you assume the expectation 0, then check probability of the value and values further (from 0), double it to get p value, and p value is less than 0.05.

a suspicious very small p value means statiscally signifcant, which means you have a nice confidence to say the 0 assumption wrong, or statiscally rejected.

Today is Sir Arthur Conan Doyle's birthday, so I would take his geneous detective for example. As Sherlock Holmes had got his bunch of evidences, the suspect could had the chance to be innocent, and the evidences could had been pure coincidences, but the probability was so low that the innocence assumption should be rejected.

However, his p value could be far less than 0.05, I think. And it's very responsible for a statician to not only say the signifancant conclusion but also give out the p value for others to judge.

*Last edited by George,Y (2006-05-22 03:15:31)*

**X'(y-Xβ)=0**

Offline

Pages: **1**