Six Sigma

Just a fun little question:

Years ago I went through Six Sigma-Black Belt training (basically applying statistics to processes)

I had the hardest time getting to the bottom of the significance of the sigma.

“Sigma is the Standard Deviation”

“The Standard Deviation is about 68% of the population”

“The Standard Deviation is this formula s=@#$+!#$^@&#@%-blahblahblah (whatever it is)”

Being a “visual” guy I got a little closer to what I was looking for with “It’s a point on the standard bell curve”

So “The Point” – Aside from all the stuff above, do you know what the significance of “That Point” is?

I know it is not just some random point that a mathematician threw a dart at and then came up with an equation. There IS significance.

Do you know what it is?

  • Yes, I do. Just wondering if you do ?

 

I tried a few of those “standards” kind of things, but problems always developed with implementation. So I started focusing on that end of it. Implementation gets to be more about personalities and the psychology of project management, which is not where I expected to end up. So, no, I don’t know what the point is.

https://www.youtube.com/watch?v=UBLbdDAzNfM

A little more thinking:

I guess you can call it Trivia.

But it does have a mathematical significance (other than “one standard deviation”)

Then I could see how someone can philosophize a business significance and sell it.

And this could maybe go under General, but I Thought being “math” it would fall under Science.

@lausten

Implementation gets to be more about personalities and the psychology of project management, which is not where I expected to end up.
Yes, I found that too. Which is one reason why I don't want to be a project manager.

I’d rather deal with hardware and programs that don’t have attitude (usually)

Are you talking about Bayesian Statistics?

When you figure it out let me know.

On the surface it makes sense, until I start thinking about it, then I go down hill in a hurry.

The likelihood function for n observations from a Normal distribution is given by the product of the Normal probability densities for each sample:

https:// www. vosesoftware. com/riskwiki/BayesianestimateofthestandarddeviationofaNormaldistributionwithknownmean . php


Then it gets complicated. ?

 

?

The likelihood function for n observations from a Normal distribution is given by the product of the Normal probability densities for each sample:
That's one way to put it.

If that Normal distribution is a standard bell curve, there is a precise point on the curve where the tangent to the slope is infinite - straight up and down - The point of inflection - That is defined as “one standard deviation” It is were the curve changes from concave to convex

So I can see from a mathematical point of view where that point is of particular interest.

But a simple 6s is probably not mathematically significant to anything but humans. … unlike infinity

 

I’m not aware of any vertical tangent on a standard bell curve.

6s is about quality control, no?

Yes it is about quality control.

OK - Maybe I was being presumptuous about the tangent - but to me it seems logical that’s what it would be where the curve changes from concave to convex - the point of inflection.