Leibnitz Formulae for differentiation - Tricky question

2 replies · 3K views · last reply
thomas49th
Messages
645
Reaction score
0
Printed homework problem 10: y'' + (4x^2 + 1)y = 0 with y(0) = 0, y'(0) = 1; asks to differentiate m times via Leibnitz to show y^(m+2)(0) = -[y^(m)(0) + 4m(m-1)y^(m-2)(0)] for m ≥ 2, deduce y(x) = sum from m=0 to infinity of c_m x^(2m+1), and find the first three non-zero terms.


All I know about Leibnitz is

[tex]\frac{d^{n}}{dx^{n}}f(x) = \sum^{n}_{i=0} \left(\stackrel{n}{i}\right) g^{n-i}(x)h^{i}(x)[/tex]

and I don't understand what the answer sheet says:

Parts of it perhaps - I can see they take the 0th, 1st and 2nd derivative, but why not more? I mean why stop at 2? I know m>=2 for y^(m+2) to be +ve, but if m was 3 (which it could be) why do we not have the 3rd derivative in there aswell? Also using the combination for working out the binomal coefficients, taking the 1st derivative for example, how does [tex]\frac{(m-1)!}{(m-2)!1!} = m[/tex]? I get m-1.
Handwritten derivation: y'' = -(4x^2+1)y, Leibniz rule applied to y^(m+2) for m≥2, then Leibniz for m=1 giving y''' = -[(4x^2+1)y' + 8xy], series solution y(x) = x - x^3/6 - 23x^5/120 ...


Stumped.
 
Last edited:
thomas49th said:
Printed math problem 10: a function y(x) satisfies y'' + (4x^2 + 1)y = 0 with y(0) = 0, y'(0) = 1; differentiate m times to show y^(m+2)(0) = -[y^(m)(0) + 4m(m-1)y^(m-2)(0)], then find the first three non-zero terms of y(x) = sum of c_m x^(2m+1).


All I know about Leibnitz is

[tex]\frac{d^{n}}{dx^{n}}f(x) = \sum^{n}_{i=0} \left(\stackrel{n}{i}\right) g^{n-i}(x)h^{i}(x)[/tex]
That doesn't even make sense- unless you say f(x)= g(x)h(x) first!

and I don't understand what the answer sheet says:

Parts of it perhaps - I can see they take the 0th, 1st and 2nd derivative, but why not more? I mean why stop at 2? I know m>=2 for y^(m+2) to be +ve, but if m was 3 (which it could be) why do we not have the 3rd derivative in there aswell? Also using the combination for working out the binomal coefficients, taking the 1st derivative for example, how does [tex]\frac{(m-1)!}{(m-2)!1!} = m[/tex]? I get m-1.
Here, [itex]f(x)= 4x^2+ 1[/itex] so that [itex]f'(x)= 8x[/itex], [itex]f''(x)= 8[/itex] and [itex]f'''(x)= 0[/itex]. It is the fact that the third derivative (and all higher) derivatives of f are 0 that is important, not derivatives of y. Indeed, there is no reason to think that y is a polynomial at all,


Handwritten derivation of a power series solution to y'' = -(4x^2+1)y using Leibniz rule, with initial conditions y(0)=0, y'(0)=1, ending in the series y(x) = x - x^3/6 - 23x^5/120 + ...


Stumped.
 
HallsofIvy said:
Here, [tex]f(x)= 4x^{2} + 1[/tex] so that , and . It is the fact that the third derivative (and all higher) derivatives of f are 0 that is important, not derivatives of y. Indeed, there is no reason to think that y is a polynomial at all,

How come f(x) equals the coefficient of y?