My Math Forum A Maclaurin expansion of arccos(-z) using only non-negative coefficients

 Calculus Calculus Math Forum

 August 2nd, 2016, 08:30 AM #1 Newbie     Joined: Aug 2016 From: Greece Posts: 3 Thanks: 0 A Maclaurin expansion of arccos(-z) using only non-negative coefficients I am trying to express the function $f(z) = \arccos(-z), \lvert z \lvert \leqslant 1$ with a Maclaurin expansion using only non-negative coefficients in a compact form as $$f(z) = \sum_{n=0}^\infty a_nz^n, a_n\geqslant0 \;\;\;(1)$$ From trigonometric identities and standard expansion of arccos/arcsin in infinite series with binomial coefficients it seems easy as $$\arccos(x) = \frac{\pi}{2}-\arcsin(x) = \frac{\pi}{2} - \sum_{k=0}^\infty\frac{(2k)!}{2^{2k}(k!)^2(2k+1)}x ^{2k+1}$$ while substituting $x$ with $-z,$ yields $$f(z)=\frac{\pi}{2} + \sum_{k=0}^\infty\frac{(2k)!}{2^{2k}(k!)^2(2k+1)}z ^{2k+1}$$ However, even if this expansion is using only non-negative coefficients is not exactly as in (1). *It was suggested to me to use the transformation* $n=2k+1 \Rightarrow$ $$f(z)=\frac{\pi}{2} + \sum_{n=1}^\infty\frac{(n-1)!}{2^{n-1}(\frac{(n-1)!}{2!})^2n}z^{n} = \frac{\pi}{2} + \sum_{n=1}^\infty\frac{2^3}{2^{n}(n-1)!n}z^{n} = \frac{\pi}{2} + \sum_{n=1}^\infty\frac{2^{3-n}}{n!}z^{n}$$ which is closer but not the same A little more help plz
 August 2nd, 2016, 03:57 PM #2 Global Moderator   Joined: May 2007 Posts: 6,641 Thanks: 625 It looks to me there is no problem. All the coefficients in the expression are non-negative. It so happens that for even powers of z, they are = 0. n=2k+1 description is faulty, since n assumes only odd values. Thanks from gepgep
August 4th, 2016, 02:31 AM   #3
Newbie

Joined: Aug 2016
From: Greece

Posts: 3
Thanks: 0

Quote:
 Originally Posted by mathman It looks to me there is no problem. All the coefficients in the expression are non-negative. It so happens that for even powers of z, they are = 0. n=2k+1 description is faulty, since n assumes only odd values.

one nitpick: Shouldn't pi/2 somehow be inside the sum (possibly in a different sum via some other transformation) in order to look like the closed form of (1)?

August 4th, 2016, 03:57 PM   #4
Global Moderator

Joined: May 2007

Posts: 6,641
Thanks: 625

Quote:
 Originally Posted by gepgep thanks for your answer especially for the n=k+1 tip one nitpick: Shouldn't pi/2 somehow be inside the sum (possibly in a different sum via some other transformation) in order to look like the closed form of (1)?
$\displaystyle a_0=\frac{\pi}{2}$.

### summation of arccos(n )-arccos(n)

Click on a term to search for related topics.
 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post elviro Calculus 5 January 13th, 2013 04:31 PM daigo Algebra 3 June 30th, 2012 08:06 AM bilano99 Calculus 3 May 3rd, 2012 10:19 AM TsAmE Algebra 8 August 9th, 2010 09:47 AM gulf56 Algebra 4 December 8th, 2009 07:26 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top