Skip to main content

numerics - Numerical error in Mathieu functions



Consider the MathieuCharacteristicA function, which is a piecewise function according to the documentation. The discontinuity happens at integer number.


With[{V0 = -1}, 
 Plot[MathieuCharacteristicA[κ, V0], {κ, -2.5, 2.5}]]

enter image description here


Consider the point approaching k=2 from the left side , and plot the Mathieu funtions near that point.


ParallelTable[
 Plot[Evaluate@
   With[{V0 = -1, κ = 2 - ϵ}, 
    Re@MathieuC[MathieuCharacteristicA[κ, V0], V0, 

      z]], {z, -10, 10}, PlotRange -> All, 
  ImageSize -> Medium], {ϵ, {10^-8, 15/10*10^-8, 18/10*10^-8,
    2*10^-8}}]

enter image description here


We see that from points k=2-10^-8 , k=2-1.5*10^-8 to points k=1.8*^-8, k=2*^-8, there are big discontinouity. Why does this big discontinuity happen in the Mathieu function, even we are still away from the piecewise point? Which result is correct?


Moreover, as I increase the working precision, the results changes. Which results should I trust?


ParallelTable[
 Plot[Evaluate@
   With[{V0 = -1, κ = 2 - ϵ}, 

    Re@MathieuC[MathieuCharacteristicA[κ, V0], V0, 
      z]], {z, -10, 10}, PlotRange -> All, ImageSize -> Medium, 
  WorkingPrecision -> 50], {ϵ, {10^-8, 15/10*10^-8, 
   18/10*10^-8, 2*10^-8}}]

enter image description here


Update:


More strange behavior


NLimit[
 Re@MathieuC[MathieuCharacteristicA[κ, -1], -1, 

   0], κ -> 2, Direction -> 1, WorkingPrecision -> 100]
(*
 0.000026560352729499428275267693547091828644960849846890155742135607985075453865741662994877041
*)


N[
 Table[Re@MathieuC[MathieuCharacteristicA[2 - ϵ, -1], -1, 
    0], {ϵ, {10^-6, 10^-8, 10^-10, 10^-20, 10^-40, 10^-60, 
    10^-100}}], 100]

(* \
{9.375519741470728355592491183508603286638427801561870220416306315833951776806837902179623867179198570*10^-6, 
 9.375519742493871990285719573456924995106820123921565403331967009687923231481580841077469030562191305*10^-8, 
 9.375519742493974304649207591451232553957148872416907065490732452953108780592926489536671069329119067*10^-10, 
 9.375519742493974314881667185336397875216528878100913700967589255793432748631135267810942389900367393*10^-20, 
 9.375519654864253585910474819580416042344587293945440367769556457939085038181962308922619162634748808*10^-40,
1.114388591781733115021002428520876171768008184684161143978162399223107768160400228809114697590700165,
1.114388591781733115021002428520876171768008184684161143978162399223107768160400228809114697590700165} *)

What's the correct limit for k->2 ?




Answer



If we compare N on the exact values with the MachinePrecision values, we see that the second two graphs (of the first quartet) look correct and the first two are wrong.


Block[{z = 0},
 Table[With[{V0 = -1, κ = 2 - ϵ}, 
   Re@MathieuC[MathieuCharacteristicA[κ, V0], V0, z]], {ϵ, {15/10*10^-8, 18/10*10^-8}}]
 ]
N[%, 6]
(*
{MathieuC[MathieuCharacteristicA[399999997/200000000, -1], -1, 0], 
 MathieuC[MathieuCharacteristicA[999999991/500000000, -1], -1, 0]}


{1.4063279613740616126811177594156`6.*^-7, 1.6875935536488557009743434482017`6.*^-7} 
*)

Block[{z = 0.},
 Table[With[{V0 = -1, κ = 2 - ϵ}, 
   Re@MathieuC[MathieuCharacteristicA[κ, V0], V0, z]], {ϵ, {15/10*10^-8, 18/10*10^-8}}]
 ]
(*
{1.11439, 1.7027*10^-7}

*)

As the OP showed, this can be seen in the plots, if the WorkingPrecision is set high enough. Clearly, one should trust the second quartet of plots if one is going to trust Mathematica at all. N[expr, n] will report the answer with a precision that is supposed to be correct.



Edit


As @acl has observed Mathieu functions are difficult functions numerically. I would expect it to be even more difficult near a singular point of MathieuCharacteristicA[κ, -1], where a little round-off error causes a discontinuous jump.


The following seem entirely consistent with the left-hand limit being zero, which is at the same time not inconsistent with the OP's evaluation of NLimit.


N@Block[{$MaxExtraPrecision = 500},
  NLimit[Re@MathieuC[MathieuCharacteristicA[κ, -1], -1, 0],
    κ -> 2, Direction -> 1, WorkingPrecision -> 300, Terms -> 50]

  ]

N[Re@MathieuC[MathieuCharacteristicA[2, -1], -1, 0], 10]

N@Block[{$MaxExtraPrecision = 500},
  NLimit[Re@MathieuC[MathieuCharacteristicA[κ, -1], -1, 0],
    κ -> 2, Direction -> -1, WorkingPrecision -> 300, Terms -> 50]
  ]
(*
-1.49047*10^-45

0.8157268391
1.15361
*)

The inconsistencies observed by the OP seem to be due to round-off error. I suppose one complaint is that Mathematica issues no warnings in evaluating the OP's examples, especially the ones involving N applied to an exact value.


Comments

Post a Comment

Popular posts from this blog

functions - Get leading series expansion term?

Given a function f[x] , I would like to have a function leadingSeries that returns just the leading term in the series around x=0 . For example: leadingSeries[(1/x + 2)/(4 + 1/x^2 + x)] x and leadingSeries[(1/x + 2 + (1 - 1/x^3)/4)/(4 + x)] -(1/(16 x^3)) Is there such a function in Mathematica? Or maybe one can implement it efficiently? EDIT I finally went with the following implementation, based on Carl Woll 's answer: lds[ex_,x_]:=( (ex/.x->(x+O[x]^2))/.SeriesData[U_,Z_,L_List,Mi_,Ma_,De_]:>SeriesData[U,Z,{L[[1]]},Mi,Mi+1,De]//Quiet//Normal) The advantage is, that this one also properly works with functions whose leading term is a constant: lds[Exp[x],x] 1 Answer Update 1 Updated to eliminate SeriesData and to not return additional terms Perhaps you could use: leadingSeries[expr_, x_] := Normal[expr /. x->(x+O[x]^2) /. a_List :> Take[a, 1]] Then for your examples: leadingSeries[(1/x + 2)/(4 + 1/x^2 + x), x] leadingSeries[Exp[x], x] leadingSeries[(1/x + 2 + (1 - 1/x...
Post a Comment
Read more

How to thread a list

I have data in format data = {{a1, a2}, {b1, b2}, {c1, c2}, {d1, d2}} Tableform: I want to thread it to : tdata = {{{a1, b1}, {a2, b2}}, {{a1, c1}, {a2, c2}}, {{a1, d1}, {a2, d2}}} Tableform: And I would like to do better then pseudofunction[n_] := Transpose[{data2[[1]], data2[[n]]}]; SetAttributes[pseudofunction, Listable]; Range[2, 4] // pseudofunction Here is my benchmark data, where data3 is normal sample of real data. data3 = Drop[ExcelWorkBook[[Column1 ;; Column4]], None, 1]; data2 = {a #, b #, c #, d #} & /@ Range[1, 10^5]; data = RandomReal[{0, 1}, {10^6, 4}]; Here is my benchmark code kptnw[list_] := Transpose[{Table[First@#, {Length@# - 1}], Rest@#}, {3, 1, 2}] &@list kptnw2[list_] := Transpose[{ConstantArray[First@#, Length@# - 1], Rest@#}, {3, 1, 2}] &@list OleksandrR[list_] := Flatten[Outer[List, List@First[list], Rest[list], 1], {{2}, {1, 4}}] paradox2[list_] := Partition[Riffle[list[[1]], #], 2] & /@ Drop[list, 1] RM[list_] := FoldList[Transpose[{First@li...
Post a Comment
Read more

front end - keyboard shortcut to invoke Insert new matrix

I frequently need to type in some matrices, and the menu command Insert > Table/Matrix > New... allows matrices with lines drawn between columns and rows, which is very helpful. I would like to make a keyboard shortcut for it, but cannot find the relevant frontend token command (4209405) for it. Since the FullForm[] and InputForm[] of matrices with lines drawn between rows and columns is the same as those without lines, it's hard to do this via 3rd party system-wide text expanders (e.g. autohotkey or atext on mac). How does one assign a keyboard shortcut for the menu item Insert > Table/Matrix > New... , preferably using only mathematica? Thanks! Answer In the MenuSetup.tr (for linux located in the $InstallationDirectory/SystemFiles/FrontEnd/TextResources/X/ directory), I changed the line MenuItem["&New...", "CreateGridBoxDialog"] to read MenuItem["&New...", "CreateGridBoxDialog", MenuKey["m", Modifiers-...
Post a Comment
Read more