So you say that if I take a couple of data points that are accurate, and have something to do with what I want to estimate, and then I’ll make up some arbitrary factors with which I’ll mesh the initial data into the required data I’ll end up with an estimation near the magnitude of the actual value.
Hmm. I may sound stupid, but that’s only a case when you are a pretty good guesser and you don’t have a bias to over/underestimate things. I think (and I’m quite good in math) this approximation is only possible if your factors are quite close (1 order of magnitude) and are randomly almost half of them under- and the others are overestimates.
So you would get a guess 1 order of magnitude close to something if you first do 6 other guesses with that proximity. Hmmm. Now that sounds a bit more tedious than just guessing on your gut and as much useful.
If I want to take a closer look at this using probability theory:
If you do like 8 random guesses, and take their product you will get a middle value most probably. A middle value you wanted in the first place.
I don’t want to sound smart, because I’m not, but this is just another example for the Take 13 way too complicated numerical calculations and a seed. Do random amount of these on the seed in each cycle and do random number of cycles. false pseudo-random generator.
The BLS maintains data on this, and I have access to statistical software with 2009 data… for the DOT occupational title 730.361-010 Piano Tuner in Cook County, IL, the software beleives there to 25 currently employed piano tuners, and 8 additional labor market job openings this year.
It also states their Mean Average Earnings as $43930.29/yr, with an SAe of 0.28
If you register on the site (go to home page to register) then it will let you save your work as you go. There are other tools as well, please have a look at the services tab on the home page