Mathematics desk
< December 31	<< Dec \| January \| Feb >>	January 2 >

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.

January 1

What is the difference between Mathematics Q&A on Stack Exchange and Wikipedia Reference Desk

I'm doing a research project about free Q&A and I would like to know your opinion about these different if you have. Thank you 133.19.15.12 (talk) 08:16, 1 January 2016 (UTC)Yousef[reply]

Please see the notice at the top of this page: We don't answer requests for opinions, predictions or debate. Two obvious factual differences that come to mind are (1) the volume of questions on math.stackexchange.com is much higher than on the RefDesk; and (2) StackExchange has a formal system that lets users vote on the quality/accuracy of answers. AndrewWTaylor (talk) 10:42, 1 January 2016 (UTC)[reply]

You need to ask a mathematical question. Non-mathematical question should be asked on the Humanities page. 110.22.20.252 (talk) 12:19, 1 January 2016 (UTC)[reply]

If 'you are doing a research project' then it is your task to find and point out the differences. Do your homework yourself. --CiaPan (talk) 12:52, 3 January 2016 (UTC)[reply]

Jeez folks no need to bit off his head. Those responses will make an interesting data point though! Dmcq (talk) 14:56, 3 January 2016 (UTC)[reply]

Interpolating sine

Using the error expression of interpolation polynomial one obtains that we can approximate the function sin on the interval $[0,2^{n}]$ with an error bounded by $\approx {\frac {1}{n!}}$ using n points (in Chebyshev nodes).

My question is: With n points we get an interpolation polynomial of degree $\leq n$ , so it has up to n roots, but sine has $\approx 2^{n}$ roots in this interval. So, how could this come true? (or: How could we approximate it that well?) Notes:

The notation of "approximation" above should be understood as asymtotic notation.
The error expression could be found here: https://en.wikipedia.org/wiki/Chebyshev_nodes (the last expression)

213.8.204.34 (talk) 21:14, 1 January 2016 (UTC)[reply]

Check the error bound. There's a factor of ((b-a)/2)ⁿ which in this case would be 2^n(n-1). --RDBury (talk) 04:57, 2 January 2016 (UTC)[reply]

Oopss... My subtitution was wrong.. Thanx :) 213.8.204.22 (talk) 06:44, 2 January 2016 (UTC)[reply]