Percentages

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Percentages
Message posted by John Williams (via 213.122.223.194) on February 26, 2001 at 5:33 PM (ET)

I have 5 variables, each with a given percentage output. What is the best way to combine them in order to give a single output for a prediction? I cannot be absolutely sure that the variables are totally independent.

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Percentages
Message posted by nancy diehl (via 129.176.179.23) on February 28, 2001 at 2:21 PM (ET)

Assuming that your percentages are not independent.... then the
result would be the product of each of your five percentages.
For example, if I have a process to produce some parts and I start
with 100 parts, go to step 1 of the process and it yields 80 good and 20 bad,
then that step has an 80% yield. When I take these 80 good to step 2 in the process,
and now I have 60 good/20 bad (or 75%), then the first time yield of this
process is .80 x .75 = .60 or 60% (60/100).
Again, if the steps are dependent then the
overall percentage is the product of all the percentages.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.