Wikipedia:Reference desk/Archives/Mathematics/2012 April 11

Welcome to the Wikipedia Mathematics Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

April 11

Millions, billions, trillions?

A document from UNCTAD, reporting on trade between EU and Africa, gives this number for 2009 EU imports from Africa, in '$m'. 231,681.000 . In the American system I (try) to understand, there would only be commas, and no decimal points, unless we were denoting cents, or eurocents, or something small. The mix of 231 comma 681 decimal 3 zeroes has me a little disconeffulated. I suppose it could be 231 million, million, and if so, is that 2.3 trillion? or 23 billion? Thanks if you can help. — Preceding unsigned comment added by 83.84.219.135 (talk) 14:39, 11 April 2012 (UTC)

That is confusing. I believe it means $1 million × 231681, which would be 231 billion dollars ($231,681,000,000). This seems like a plausible amount of money given the context. Having 3 decimal places is not unreasonable since it's measuring fractions of a million dollars rather than fractions of a dollar (where we would only expect 2 decimal places), although it's a little weird that they're included when they're all zero. 231 million million is equal to 231 trillion which is much too high. Similarly 231 million dollars would be too low. Rckrone (talk) 15:01, 11 April 2012 (UTC)

It's especially weird considering that adding those zeros implies that the figures are significant but it's highly unlikely that the actual trade amount is such an exact round number. 1.155.25.201 (talk) 05:13, 12 April 2012 (UTC)

Agreed. I'd write that as $231.681 billion, if I felt the need for such high accuracy. I'd write it as $232 billion, if not. (And I bet there's like a 10% margin of error in that calculation, anyway, making it rather absurd to include all those digits.) StuRat (talk) 03:55, 13 April 2012 (UTC)

Well, accounting is not quite like physics. Assets are computed to the penny, even when including intangibles like "good will" that no one really knows how to value accurately. I couldn't say whether this report would do something like that, but it's not out of the question.

As for significant figures, just because you see three zeroes in a row shouldn't make you assume they're not significant. One time in a thousand, three significant figures will all be zero, so it's not particularly rare. I admit I don't have an explanation that would make the first three zeroes significant and the latter three not. --Trovatore (talk) 09:56, 13 April 2012 (UTC)

To me, including that many digits beyond the actual accuracy is a form of deception, to try to convince the readers the figures are more precise than they really are. StuRat (talk) 16:29, 13 April 2012 (UTC)

Yes, this is the standard view in the sciences. As I said, it's different in accounting. If I have $36.27 in my bank account and ten million dollars worth of good will, that's $10,000,036.27. I think the reason is that otherwise people might be able to skim money without it showing up on the balance sheet. --Trovatore (talk) 21:15, 13 April 2012 (UTC)

It seems to me that adding "apples and oranges" like that is bound to get you into trouble. In your example, I could take the cash and argue that they just increased their good will at least that much by making a customer happy. So could everyone else. They will end up with a very popular, and very bankrupt company. StuRat (talk) 21:42, 13 April 2012 (UTC)

Well, accounting is a very old discipline, and I imagine that they have ways of dealing with such things. I don't know what they are as I've never been very interested in accounting. But somebody has to do it. --Trovatore (talk) 21:59, 13 April 2012 (UTC)

That also reminds me of the kid in the museum who asks the guard how old the dinosaur is. He answers 200,000,007 years old, since it was 200,000,000 years old when he started working there, 7 years back. StuRat (talk) 21:45, 13 April 2012 (UTC)

47.38256% of people don't understand mathematics well enough to get this stuff right. (Oh, and 70% of statistics are made up on the spot.) HiLo48 (talk) 23:42, 13 April 2012 (UTC)

Use of anova/ F-tests with count data

Question posted on Science desk. Thankyou. IBE (talk) 19:50, 11 April 2012 (UTC)