Evaluation of ζ(2) by the definition (sort of) [HP-42S & HP-71B]
|
10-31-2021, 03:40 PM
(This post was last modified: 11-06-2021 03:00 PM by Albert Chan.)
Post: #13
|
|||
|
|||
RE: Evaluation of ζ(2) by the definition (sort of) [HP-42S & HP-71B]
Another approach, getting ζ(2) with alternating series.
Σ(all terms) = Σ(even terms) * 4 Σ(odd terms) = Σ(all terms) - Σ(even terms) = Σ(even terms) * 3 Σ(odd terms) - Σ(even terms) = Σ(even terms) * 2 Let s = (-1)^n, x=n*n+n+1, T's = triangular number ζ(2) ≈ (2 - 2/2^2 + 2/3^2 - ... + s*2/n^2) - s/(x - 1/(x+4*T1 - 2^4/(x+4*T2 - 3^4/(x+4*T3 - 4^4/(x+4*T4 - ... Note: Tk - Tk-1 = k*(k+1)/2 - (k-1)*k/2 = k zeta2_alt(n) is computationally less expensive than zeta2(n) Code: function zeta2_alt(n) Update: revert zeta2_alt(n) with b=0, to fairly compare against zeta2(n) (also, b=0) zeta2_alt(n) is not as good, less accurate by about 2/5 of a decimal digit. lua> err = function(z2) return abs(pi*pi/6-z2) end lua> for n=1,6 do print(n, log10(err(zeta2_alt(n))/err(zeta2(n)))) end 1 0.3872388516915953 2 0.40466274755687004 3 0.4105758432666672 4 0.4132661850460007 5 0.41470949038470145 6 0.41521536217708266 |
|||
« Next Oldest | Next Newest »
|
User(s) browsing this thread: 1 Guest(s)