I am asked to find all solution to the equation: $$y'= \left( \begin{array}{cc} 13 & 12 \\ 12 & 13 \end{array} \right) y+ \left(\begin{array}{c} x\\ 0 \end{array} \right)$$ No initial condition is specified.

My working so far: If we write the DE as $y'=Ay+b(x)$ then a solution to the homogeneous DE $y'=Ay$ is given by: $$y_h=e^{A(x-x_0)}y_0$$ where $y_0$ is a vector containing the arbitrary initial conditions. Because no initial conditions are specified could I just assume that $x_0=0$? I then probably need to use variation of parameters to find a general solution but I am unsure how to do this for a system of linear differential equations. Also, how do I calculate $e^{A(x-x_0)}$, is $A$ diagonalizable? If so I could calculate it. My linear algebra is not too great so I got stuck here too. Any help would be appreciated. Thanks!


Do you know how to find the Eigenvalues and Eigenvectors of a matrix?

To find the eigenvalues, you solve for the roots of the characteristic polynomial using $(\mathbf{A} - \lambda \mathbf{I}) = 0$?

For your problem, you have

$$ \mathbf{A} = \begin{bmatrix} 13 & 12\\ 12 & 13 \end{bmatrix} $$

We form $(\mathbf{A} - \lambda \mathbf{I}) = 0$, so

$$ (\mathbf{A} - \lambda \mathbf{I}) = \begin{bmatrix} 13- \lambda & 12\\ 12 & 13 - \lambda \end{bmatrix} = 0$$

$$(13 - \lambda)^{2} - 144 = 0, \text{so}, \lambda_{1,2} = 1, 25$$

To find the corresponding eigenvectors, you substitute each distinct (if they are not distinct, other approaches are needed) eigenvalue into and by solving $(\mathbf{A} - \lambda_i \mathbf{I})\mathbf{x} = 0$.

For this example, you would get:

$$\lambda_1 = 1, v1 = (1, 1)$$

$$\lambda_2 = 25, v2 = (-1, 1)$$

You could also approach this using the Jordan Normal Form and many other ways too.

Can you take it from here?


  • $\begingroup$ Thanks, yeah I know about the eigenvalues and eigenvectors. So my homogeneous solution would be $y_h=c_1(1,1)^T+c_2(-1,1)^T$ right? How do I go from this to the general solution to the non-homogeneous solution? thx $\endgroup$ – Slugger Jan 25 '13 at 16:30
  • $\begingroup$ See Section 3. You are forgetting the exponential terms. Regards $\endgroup$ – Amzoti Jan 25 '13 at 16:33
  • $\begingroup$ How'd this go with no upvotes for so long...I made one +1 dent in that! ;-) $\endgroup$ – Namaste May 6 '13 at 1:47
  • $\begingroup$ @amWhy: Not even after a followup from the OP! :-) Thanks my friend. $\endgroup$ – Amzoti May 6 '13 at 1:50

Take $x_0=0$ and $y_0=(c_1,c_2)^t\in\mathbb{R}^2$. As $A$ is real and symmetric, is diagonalizable in $\mathbb{R}$.


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.