0
$\begingroup$

I have the following linear equation system: $$ (S)=\left\{ \begin{array}{ll} a2008 + b = 34.564\\ a2009 + b = 35.104\\ a2010 + b = 35.481\\ a2011 + b = 36.891\\ a2012 + b = 37.331\\ a2013 + b = 38.061\\ a2016 + b = 39.071\\ a2018 + b = 39.345 \end{array} \right. $$ and the corresponding augmented matrix $$ \mathbf{M} = \left[ \begin {array}{ccc} 2008&1& 34.564\\ {}2009&1& 35.104\\ {}2010&1& 35.481\\ {}2011&1& 36.891\\ {}2012&1& 37.331\\ {}2013&1& 38.061\\ {}2016&1& 39.071\\ {}2018&1& 39.345\end {array} \right] $$ After bringing this to reduced row-echelonform it is evident that rank $\mathbf{M}$ > rank $\mathbf{A}$ and that the linear equation system therefore is inconsistent - note that $\mathbf{A}$ is the coefficient matrix.

However, bringing the augmented matrix to RREF takes time, and for some reason I was able to figure that it was consistent before performing any row operations. I think it is because that there are all ones in the second column but it does not seem like a satisfactory argument/explanation.

So how would you argue that this linear equation system clearly has no solution (is inconsistent) without having the calculate anything?

$\endgroup$
  • $\begingroup$ That is no equation system, it is a matrix. Maybe write down the equation? $\endgroup$ – mathreadler Jul 26 '19 at 10:35
  • $\begingroup$ It is the augmented matrix of a linear equation system as I wrote? $\endgroup$ – npp Jul 26 '19 at 10:58
  • $\begingroup$ But yea, I can also write the linear equation system in the question, even though it should not be necessary when you already have the augmented matrix for that system :)) $\endgroup$ – npp Jul 26 '19 at 10:59
  • 2
    $\begingroup$ Without any matrices or any fancy stuff, just look at the differences of the different rows in the first equation. On the left-hand side you always get $a$ (or a multiple of $a$), but what do you get on the right-hand side? Based on this, the right-hand side difference should always be the same ... Just take a look at the first three rows, for example. $\endgroup$ – Matti P. Jul 26 '19 at 11:04
  • $\begingroup$ In addition to what @MattiP. says: I generally don't use RREF reduction. You can use the fact that if there is a non-null minor $M$ of order $k$ and all minors of order $k+1$ obtained adding one row and one column to $M$ are null, then the rank of the matrix is precisely $k$. $\endgroup$ – dfnu Jul 26 '19 at 11:10
2
$\begingroup$

Since you have only two independent variables $a$ and $b$, any two linear equations in $a$ and $b$ will have a unique solution (as long as they are independent). So you can pick two pairs of equations from your list, solve for $a$ and $b$, and see if the solutions are the same.

In this case it is particular easy to see that the equations in $S$ are not consistent:

$2008a + b = 34.564 \\ 2009a + b = 35.104 \\ \Rightarrow a = 35.104 - 34.564 = 0.54$

$2009a + b = 35.104 \\ 2010a + b = 35.481 \\ \Rightarrow a = 35.481 - 35.104 = 0.377$

$\endgroup$
1
$\begingroup$

It really doesn't take much effort to reduce $\mathbf M$.

$$ \left[ \begin {array}{ccc} 2008 & 1 & 34.564 \\ 2009 & 1 & 35.104 \\ 2010 & 1 & 35.481 \\ 2011 & 1 & 36.891 \\ 2012 & 1 & 37.331 \\ 2013 & 1 & 38.061 \\ 2016 & 1 & 39.071 \\ 2018 & 1 & 39.345 \end {array} \right] $$

The result of \begin{align} 2\mathbf M_2 &\to \mathbf M_2 \\ \mathbf M_2 - \mathbf M_1 &\to \mathbf M_2 \\ \mathbf M_2 - \mathbf M_3 &\to \mathbf M_2 \\ \dfrac{1}{0.163} \mathbf M_2 &\to \mathbf M_2 \\ \end{align}

is

$$ \left[ \begin {array}{ccc} 2008 & 1 & 34.564 \\ 0 & 0 & 1.000 \\ 2010 & 1 & 35.481 \\ 2011 & 1 & 36.891 \\ 2012 & 1 & 37.331 \\ 2013 & 1 & 38.061 \\ 2016 & 1 & 39.071 \\ 2018 & 1 & 39.345 \end {array} \right] $$

And we will quickly get

$$ \left[ \begin {array}{ccc} 2008 & 1 & 0 \\ 0 & 0 & 1 \\ 2010 & 1 & 0 \\ 2011 & 1 & 0 \\ 2012 & 1 & 0 \\ 2013 & 1 & 0 \\ 2016 & 1 & 0 \\ 2018 & 1 & 0 \end {array} \right] $$

$$ \left[ \begin {array}{ccc} 2008 & 1 & 0 \\ 0 & 0 & 1 \\ 2 & 0 & 0 \\ 3 & 0 & 0 \\ 4 & 0 & 0 \\ 5 & 0 & 0 \\ 8 & 0 & 0 \\ 10 & 0 & 0 \end {array} \right] $$

$$ \left[ \begin {array}{ccc} 2008 & 1 & 0 \\ 0 & 0 & 1 \\ 1 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \end {array} \right] $$

$$ \left[ \begin {array}{ccc} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \\ 0 & 0 & 0 \end {array} \right] $$

$\endgroup$

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.