3
$\begingroup$

If one knows that the solutions of $y''+ y=0$ are two functions $s(x)$ and $c(x)$, and we know that $s(0)=0$, $s'(0)=1$, $c(0)=1$, $c'(0)=0$, then how can one start to prove that $s(x+a)=s(x)c(a)+c(x)s(a)$? What is it that has to be observed in order for one to understand how to initiate the proof?

One is not supposed to use the actual trigonometric functions in this case. Here we assume that we know nothing about sin and cos.

$\endgroup$
2
  • $\begingroup$ I substituted the RHS of the identity to the 2nd derivative part of the ODE and the LHS to the 0th derivative part of the ODE and proceeded from there. Would this be a correct proof technique, given that no construction happens in this case and the ready results are used, without a derivation as such? $\endgroup$
    – sequence
    Oct 2, 2015 at 4:41
  • 1
    $\begingroup$ Check out this answer. $\endgroup$
    – Pragabhava
    Feb 12, 2016 at 14:55

4 Answers 4

3
$\begingroup$

Here's a hint: You can use the Existence and Uniqueness theorem for linear differential equations to prove many different identities.

If you need to prove an identity $y_1(x)=y_2(x)$, first prove that both $y_1(x)$ and $y_2(x)$ are solutions to the provided differential equation. Then show that $y_1(a)=y_2(a)$ and $y_1'(a)=y_2'(a)$ for a value $a$ by using the initial conditions. Because there can only be one such solution, $y_1(x)$ must equal $y_2(x)$.

You'll want to use this approach to first prove $s'(x) = c(x)$ and $c'(x) = -s(x)$ which will allow you to differentiate each side of the identity when you need it. Then you can use the same approach to prove $s(x+a)=s(x)c(a)+c(x)s(a)$.

$\endgroup$
1
  • $\begingroup$ Unfortunately this approach kinda requires you to already know what you want to prove. $\endgroup$
    – Ian
    Feb 12, 2016 at 14:27
0
$\begingroup$

HINT:

Using this,

$$y=A\cos ax+B\sin ax$$ where $A,B$ are arbitrary constants to be determined by $$s(0)=c'(0)=0,s'(0)=c(0)=1$$

$\endgroup$
4
  • $\begingroup$ Unfortunately, one is not supposed to use the actual trigonometric functions in this case. Here we assume that we know nothing about sin and cos. $\endgroup$
    – sequence
    Oct 2, 2015 at 4:45
  • $\begingroup$ @sequence, I fear, $$y=Ae^{iax}+Be^{-iax}$$ is not acceptable either $\endgroup$ Oct 2, 2015 at 4:47
  • $\begingroup$ Without using the Euler's formula, this won't help much :-\ $\endgroup$
    – sequence
    Oct 2, 2015 at 4:52
  • $\begingroup$ @labbhattacharjee I think (but I'm not pretty sure) that answer here is supposed to use Ostrogradskii-Liouville formula. Unluckily I don't have time to write answer myself :) $\endgroup$
    – Evgeny
    Oct 2, 2015 at 6:54
0
$\begingroup$

Not an answer but, by a series approach can you establish that

$$(x+a) - (x+a)^3/3! + ( x+a)^5/5! - ...$$

$$ =( x - x^3/3! + x^5/5! -...) \cdot ( 1 - a ^2/2! + a^4/4! - ... ) +$$

$$(1- x^2/2! +x^4/4!- ... )\cdot ( a - a ^3/3! + a^5/5! - ....) ? $$

$\endgroup$
-1
$\begingroup$

We use the following Theorem: [Page 82 of Simmons Differential Equations with Applications and Historical Notes] If P(x), Q(x), and R(x) be continuous functions on a closed interval [a,b]. If x0 is any point in [a. b], and if y0 and y0' are any numbers whatever, then equation y"+P(x) y'+ Q(x)y=R(x) has one and only one solution y(x) on the entire interval such that y(x0)=y0, and y'(x0)=y0'.

Y1=**s(x+a) is a solution of y"+y=0. **Y2=**s(x)c(a)+c(x)s(a) is also a solution of y"+y=0. Also, Y1(0)=s(a) and Y2(0)=s(a). Y1'(0)=s'(x+a)|x=0=s'(0)=1 and Y2'(0)=s'(x)c(a)+c'(x)s(a)|x=0 = 1. **Hence by Theorem Y1=Y2.

$\endgroup$
1
  • $\begingroup$ Right now, your answer is quite difficult to parse. Could you please reformat it, using MathJax? $\endgroup$
    – Xander Henderson
    Oct 22, 2017 at 14:58

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .