I think I made a mistake somewhere:

$$y'' -xy' - y = 0$$ given $y(0) = 1$ and $y'(0) = 0$

so we have:

$$y = \sum_{n=0}^\infty C_nx^n$$ $$y' = \sum_{n=1}^\infty nC_nx^{n-1}$$ $$y' = \sum_{n=2}^\infty n(n-1)C_nx^{n-2}$$

so subbing:

$$y' = \sum_{n=2}^\infty n(n-1)C_nx^{n-2} - \sum_{n=0}^\infty nC_nx^{n-1} - \sum_{n=0}^\infty C_nx^{n} = 0$$

$$y' = \sum_{n=0}^\infty (n+2)(n+1)C_{n+2}x^{n} - \sum_{n=0}^\infty (n+1)C_nx^n = 0$$

$$C_{n+2} = \frac{C_n}{n+2}$$

so a few terms:

$C_0 = C_0$ and $C_1 = C_1$ and $C_2 = \frac{C_0}{2}$ and $C_3 = \frac{C_1}{3}$ and $C_4 = \frac{C_2}{4} = \frac{C_0}{4 \cdot 2}$ and $c_5 = \frac{c_3}{5} = \frac{C_1}{5 \cdot 3}$

so the even terms are: $\frac{C_0}{2^n \cdot n!}$ and the odd terms are: $\frac{C_1}{1 \cdot 3 \cdot 5 \cdot (2n-1)}$

and so $$y = \sum_{n=0}^\infty \frac{c_0}{2^n \cdot n!}x^{2n} + \sum_{n=0}^\infty \frac{c_1}{1 \cdot 3 \cdot 5 \cdot (2n-1)} x^{2n-1}$$

but I'm stuck here. If $y(0) = 1$ ... doesn't the equation become 0? I feel like I've hit an impossible condition so I feel like I've made a mistake somewhere.

  • $\begingroup$ When you have substituted in, you have changed the minimum value of n from 0,1 and 2 respectively to 0 across the board. If you do that then you must account for those finite terms as well when you change to zero $\endgroup$ – W M Seath May 10 at 15:47
  • $\begingroup$ @WMSeath I don't get it... can you show me what you mean? $\endgroup$ – Jwan622 May 10 at 15:47
  • $\begingroup$ I'm going from n = 2 to n = 0 so aren't the constants taken into account already? $\endgroup$ – Jwan622 May 10 at 15:55

You did well. Note that the initial conditions $y(0)=1$ and $y'(0)=0$ simply means $C_0=1$ and $C_1=0$.

Also, noting that $C_1=0$, your final solution is $$y=\sum_{n=0}^\infty\frac{c_0}{2^n\cdot n!}x^{2n}=c_0+\sum_{n=1}^\infty\frac{c_0}{2^n\cdot n!}x^{2n}$$ which dosn't make the equation $0$.

  • $\begingroup$ which is fine since y(0) = 1 right? $\endgroup$ – Jwan622 May 10 at 16:28
  • $\begingroup$ the comment from WMseath doesn't apply right? $\endgroup$ – Jwan622 May 10 at 16:29
  • $\begingroup$ Yes, of course. $\endgroup$ – Qurultay May 10 at 16:29
  • $\begingroup$ No, you did every thing right. What you did, is what every body do. The only typo was that when you substituted $x=0$ in the final solution, you missed to note that the series started at $n=0$, which makes $x^{2n}=0$ in the case of $n=0$. $\endgroup$ – Qurultay May 10 at 16:33
  • $\begingroup$ What I mean: first apply the index $n$, then substitute value of $x$. $\endgroup$ – Qurultay May 10 at 16:34

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.