Reply to: Any maths / statistics whizzes here?

[OwlHoot]

He's got it right. By accident I'm sure. It 0.4% probability = 0.004 probability.

Fair point - Just shows, once one person spreads confusion, others can start fecking up too!

[NotAllThere]

I am looking to see if I can extrapolate some numbers based on the information set I have.

If a sample is 100% and each week 0.4% "do a thing"...

No, now you've gone too far the other way - Not 0.004, it's 0.04%

He's got it right. By accident I'm sure. It 0.4% probability = 0.004 probability.

[OwlHoot]

oh gwd I missed a 0.004 - que ritual ridicule - fair game I suppose.

No, now you've gone too far the other way - Not 0.004, it's 0.04%

[scooterscot]

oh gwd I missed a 0.004 - que ritual ridicule - fair game I suppose.

[ladymuck]

Bored here, opened excel, put 100 into cell A1
in cell B1, I put: =A1-(A1*0.004) (i.e. B1 is the remainder after subtracting 0.4% from A1)
Copy that down and it takes 1150 weeks to get down to 1 person.
And 1323 weeks to get below 0.5
By 3448 weeks you're at 0.000100388

If you want to hit six sigma, it's 3142 weeks.


But that's probably not what you're looking for.

Do it again but this time bearing in mind the original requirement.

0.4% do the thing in a week. How long to reach 100%

[NotAllThere]

Here's my working out.



Invoice is in the post.

You idiot. You've got phi-minus where there should be a phi-plus.

[DoctorStrangelove]

But where they have been quoted, they still live on.

To the end of the world or until the oil runs out?

[jamesbrown]

Has he?

Yeah, all the ones that make him look like a lying moron. Still, he left all the ones that make him look like a regular moron.

[sasguru]

I see pooperscooter has been pruning some posts from this thread without replacement.

Has he?

[WTFH]

I see pooperscooter has been pruning some posts from this thread without replacement.

But where they have been quoted, they still live on.

[jamesbrown]

I see pooperscooter has been pruning some posts from this thread without replacement.

[WTFH]

Bored here, opened excel, put 100 into cell A1
in cell B1, I put: =A1-(A1*0.004) (i.e. B1 is the remainder after subtracting 0.4% from A1)
Copy that down and it takes 1150 weeks to get down to 1 person.
And 1323 weeks to get below 0.5
By 3448 weeks you're at 0.000100388

If you want to hit six sigma, it's 3142 weeks.


But that's probably not what you're looking for.

[sasguru]

but if he used the right number he'd get a similar number to you and NAT (1897 weeks to your 1860 ish)

What about sampling with replacement? Another hint for Pooper.

[DimPrawn]

Is this a gut feel or your expert opinion?

Here's my working out.



Invoice is in the post.

[pr1]

That's not really an issue, just semantics.
However Pooper is not completely thick, he's just reduced the problem to a level that's within his capabilities.
Here's some hints Pooper: read the past posts, work out what has been done and why.

but if he used the right number he'd get a similar number to you and NAT (1897 weeks to your 1860 ish)