More six-sigma one-liners for the statistically inclined...
When unsure about what to do with the data, use simpler data.
When unsure about what to do with the data, use a giant flowchart.
When unsure about what to do with the data, test it to destruction.
When unsure about what to do with the data, formulate a good hypothesis.
Have plan B ready: formulate an alternate hypothesis, too.
If you are too confident (an alpha person), prepare for errors.
Less than 95% is underconfident.
Hal Varian quote - If you torture the data long enough, it will reveal whatever you want it to show.
Quote from my friend Suri: When unsure about data, get more data till every one is as unsure of it as you are!
Q: If Z is 1.96, what's the risk?
A: Arre beta, it's alpha.
Examples and causes -- ex and why variables.
Why are some variables called 'x' variables?
They are called 'x's' because there is an excess of them....
Equality is a null hypothesis.
Hmm, so is the alternative one-up-manship?
Fault Tree: The sacrificial plant deep in the forest, that takes all the blame.
TIM WOOD knows all about it...
Lean six-sigma: Will soon be 4.5 sigma due to prolonged dieting.
Lean six-sigma (sing to the right tune): what you want, where you want, when you want and how you want it...
Takt time: diplomatically finish things in due time.
Showing posts with label six sigma. Show all posts
Showing posts with label six sigma. Show all posts
Monday, March 22, 2010
Wednesday, February 17, 2010
Some six-sigma one-liners
These ones are for the statistically or six-sigma inclined people out there.*
Statistics: The art of taking decisions in the face of uncertainty.
In the ongoing battle between Variable data and Attribute data, Variable data is still winning. Yes or no?
Degrees of freedom: They can have any color they want, so long as it is black -- Henry Ford on colors of the Model-T
Normal distributions are all alike; every non-normal distribution is non-normal in its own way. -- Anna Karenina version of distribution functions
Examples and causes -- ex and why variables.
Elitism in causation: some causes are special, some are merely common.
* If you're inclined at 6-sigma, there is only 3.4 chances in a million that you will not be straightened out some time.
Statistics: The art of taking decisions in the face of uncertainty.
In the ongoing battle between Variable data and Attribute data, Variable data is still winning. Yes or no?
Degrees of freedom: They can have any color they want, so long as it is black -- Henry Ford on colors of the Model-T
Normal distributions are all alike; every non-normal distribution is non-normal in its own way. -- Anna Karenina version of distribution functions
Examples and causes -- ex and why variables.
Elitism in causation: some causes are special, some are merely common.
* If you're inclined at 6-sigma, there is only 3.4 chances in a million that you will not be straightened out some time.
Subscribe to:
Posts (Atom)
Posts
