Jan

12

A very interesting article reminding us that we can estimate the standard deviation of most of our data as the range/4 or (b-a)/4 where b is the highest and a is the lowest obse ration.

This leads to many interesting augmentations. You can quickly calculate the significance of a result by 4u x n to 1/2 /(range).

I was led to this result by a very interesting 1980 Box on evolutionary manufacturing operations. I was amazed that the authors were so clear and precise about the statistics, but then I saw that the author was Box and Draper, considered two of the world's leading statisticians.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search