1
$\begingroup$

We have the following LES $I$ \begin{align} 3x_1-2x_2+4x_3-x_4&=1\\ -\frac43x_2-\frac{13}3x_3+\frac13x_4&=-\frac{19}3 \end{align} with the solution set $\mathcal{L}_{I}$, which contains all solutions of $I$. Let $\mathcal{L}_2$ be some other solution, which is defined as the following: $$\mathcal{L}_2=\{(x_1,x_2,x_3,x_4)\mid(x_1,x_2,x_3,x_4)=(3,3,1,6)+\lambda_1(3,3,-1,-1)+\lambda_2(2,3,1,4),\;\lambda_1,\lambda_2\in\mathbb{R}\}$$

We want to determine whether $\mathcal{L}_2\subseteq \mathcal{L}_{I}$ or not.

Therefore, we will insert the tuples into $I$. If all equations hold for the inserted values $x_1,\cdots,x_4$, the solution set $\mathcal{L}_2$ is a subset of all solutions.: First equation holds. Moving on to the next one. We can simplify the term $$-\frac{4}3(3+3\lambda_1+3\lambda_2)-\frac{13}{3}(1-\lambda_1+\lambda_2)+\frac13(6-\lambda_1+4\lambda_2)=-\frac{19}3$$ to $$-\frac{19}3-7\lambda_2=-\frac{19}3$$ How do we go on now? Can we choose $\lambda_2=0$ and assume $\mathcal{L}_2\subseteq \mathcal{L}_I$ because all equations hold for $\lambda_2=0$ or is this in fact the sign for $\mathcal{L}_2\nsubseteq \mathcal{L}_I$?

$\endgroup$
  • $\begingroup$ What do you mean by "THE solution"? The "solution set'' for a given problem contains all solutions to the problem. $\endgroup$ – user247327 Apr 12 at 16:57
  • $\begingroup$ I wanted to express, that $\mathcal{L}_I$ is the solution set of the linear system $I$ while $\mathcal{L}_2$ is just one set which still needs to be analysed in order to know if $\mathcal{L}_2$ is a subset of the solution set of $I$ $\endgroup$ – Doesbaddel Apr 12 at 17:07
  • $\begingroup$ @user247327 I fixed some typos and changed the overall style a bit. I hope you understand my question better now. $\endgroup$ – Doesbaddel Apr 12 at 19:56
1
+50
$\begingroup$

If you choose $\lambda_1=0,\lambda_2=-1$ you get $x \in \mathcal{L}_2$ of the form $(1, 0, 0, 2)$, which does not solve the second equation, therefore $\mathcal{L}_2\nsubseteq \mathcal{L}_I$.

$\endgroup$
  • $\begingroup$ Is it a necessary condition that if you choose fixed $\lambda_1,\lambda_2$ that both equations must hold in order to be a subset of all solutions $\mathcal{L}$? If we choose $\lambda_2=0,\lambda_1\in\mathbb{R}$ both equations hold for example. $\endgroup$ – Doesbaddel Apr 14 at 11:19
  • $\begingroup$ If we choose $\lambda_2=0,\lambda_1\in\mathbb{R}$ both equations hold for example. Therefore, $\mathcal{L}_2\subseteq\mathcal{L}$? I don't understand that $\endgroup$ – Doesbaddel Apr 14 at 11:25
  • $\begingroup$ $\mathcal{L}_I$ is a set of all solutions of the system of equations, which you have defined above, right? So any $x \in \mathcal{L}_2$ must also a solution of this system (i.e. of both equations) in order $\mathcal{L}_2 \subseteq \mathcal{L}_I$ to be true. $\endgroup$ – Vítězslav Štembera Apr 15 at 9:28
  • $\begingroup$ Yeah, that's right. That's why I can't pick a fixed $\lambda$ because that wouldn't be any $x\in\mathcal{L}_2$? $\endgroup$ – Doesbaddel Apr 15 at 13:41
  • 1
    $\begingroup$ If you want to prove $\mathcal{L}_2\subseteq \mathcal{L}_I$ you have to prove it for all $x \in \mathcal{L}_2$, if you want to prove $\mathcal{L}_2\nsubseteq \mathcal{L}_I$ you can pick only one $x \in \mathcal{L}_2$ and show that $x$ does not belong to $\mathcal{L}_I$. $\endgroup$ – Vítězslav Štembera Apr 15 at 14:05

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.