7
$\begingroup$

I need to find $f(x,n)$ in the interval [0,1] defined by recursion,

\begin{equation} \frac{d f(x,n+1)}{dx} = f(x,n) \end{equation}

with boundary conditions $f(0,n+1) = 1$ and $f(x,0)=1+ x $

Using DSolve I can get integral of $f(x,0)$

  DSolve[{z'[x] == 1+x}, z[x], x]

which gives,

 {{z[x] -> x + C[1] + (x^2)/2}}

but how can I also specify boundary values in the equation?

$\endgroup$
6
$\begingroup$
 DSolve[{z'[x] == 1+x, z[0]==1}, z[x], x]

But it is not clear what you mean by "you need to find $f(x, n).$" You mean, a closed form for any $n$? Or?

In any case, to get it for any fixed $n,$ you can use:

ds[foo_] := (z[x] /. DSolve[{z'[x] == foo, z[0] == 1}, z[x], x])[[1]]
Nest[ds, 1+x, n]

For example:

Nest[ds, 1 + x, 50]

Gives you: $$ \frac{x^{51}+51 x^{50}+2550 x^{49}+124950 x^{48}+5997600 x^{47}+281887200 x^{46}+12966811200 x^{45}+583506504000 x^{44}+25674286176000 x^{43}+1103994305568000 x^{42}+46367760833856000 x^{41}+1901078194188096000 x^{40}+76043127767523840000 x^{39}+2965681982933429760000 x^{38}+112695915351470330880000 x^{37}+4169748868004402242560000 x^{36}+150110959248158480732160000 x^{35}+5253883573685546825625600000 x^{34}+178632041505308592071270400000 x^{33}+5894857369675183538351923200000 x^{32}+188635435829605873227261542400000 x^{31}+5847698510717782070045107814400000 x^{30}+175430955321533462101353234432000000 x^{29}+5087497704324470400939243798528000000 x^{28}+142449935721085171226298826358784000000 x^{27}+3846148264469299623110068311687168000000 x^{26}+99999854876201790200861776103866368000000 x^{25}+2499996371905044755021544402596659200000000 x^{24}+59999912925721074120517065662319820800000000 x^{23}+1379997997291584704771892510233355878400000000 x^{22}+30359955940414863504981635225133829324800000000 x^{21}+637559074748712133604614339727810415820800000000 x^{20}+12751181494974242672092286794556208316416000000000 x^{19}+242272448404510610769753449096567958011904000000000 x^{18}+4360904071281190993855562083738223244214272000000000 x^{17}+74135369211780246895544555423549795151642624000000000 x^{16}+1186165907388483950328712886776796722426281984000000000 x^{15}+17792488610827259254930693301651950836394229760000000000 x^{14}+249094840551581629569029706223127311709519216640000000000 x^{13}+3238232927170561184397386180900655052223749816320000000000 x^{12}+38858795126046734212768634170807860626684997795840000000000 x^{11}+427446746386514076340454975878886466893534975754240000000000 x^{10}+4274467463865140763404549758788864668935349757542400000000000 x^9+38470207174786266870640947829099782020418147817881600000000000 x^8+307761657398290134965127582632798256163345182543052800000000000 x^7+2154331601788030944755893078429587793143416277801369600000000000 x^6+12925989610728185668535358470577526758860497666808217600000000000 x^5+64629948053640928342676792352887633794302488334041088000000000000 x^4+258519792214563713370707169411550535177209953336164352000000000000 x^3+775559376643691140112121508234651605531629860008493056000000000000 x^2+1551118753287382280224243016469303211063259720016986112000000000000 x+1551118753287382280224243016469303211063259720016986112000000000000}{155111875328738 2280224243016469303211063259720016986112000000000000} $$

$\endgroup$
  • 5
    $\begingroup$ Makes a nice looking vase! :) $\endgroup$ – Michael E2 Sep 14 '14 at 20:19
  • $\begingroup$ @MichaelE2 my thoughts exactly... $\endgroup$ – Igor Rivin Sep 14 '14 at 22:53
4
$\begingroup$

Pushing @Igor's one step further:

ds[foo_] := (z[x] /. DSolve[{z'[x] == foo, z[0] == 1}, z[x], x])[[1]]
f = FindSequenceFunction@Apart@NestList[ds, 1 + x, 10]      

(* (E^x Gamma[1 + #1, x])/Gamma[1 + #1] & *)

Apart@Nest[ds, 1 + x, 3] == Apart@FunctionExpand@f[4]
(* True*)
$\endgroup$
  • 2
    $\begingroup$ Cool! I was not aware of this (FindSequenceFunction) at all, though I suppose in this case Apart was the real key, since once you do that, it is clear you get the partial sums of $\exp(x)$... $\endgroup$ – Igor Rivin Sep 14 '14 at 22:53
  • 1
    $\begingroup$ @IgorRivin Yup. FindSequenceFunction[] sometimes work, but don't expect too much from it. Anyway I only used your nice answer. $\endgroup$ – Dr. belisarius Sep 14 '14 at 23:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.