crisis of identity (matrix transformations)

Hi folks,

I read the following example in my maths book and feel the authors have not explained what they are doing very clearly. We have to find the equations of any straight lines that are invariant under the following transformation:



Where x' and y' are the image points of x and y. Any straight line will be of the form y = mx + c and any point on the line will have coordinates of the form (k, mk + c).

Thus



=



Now if y = mx + c is invariant, then this point (x', y') must also lie on y = mx + c

i.e.

giving: (equation 1).

This equation must be true for all values of k (i.e. equation 1 is an identity in k). Thus:

and

So: for those who have followed so far, I have to ask, how is equation 1 split into two parts? k is a variable and I know the difference between an equation and an identity (one is for all k, the other only particular values of k) but I don't understand the theory behind the way equation 1 suddenly becomes two equations both equal to zero. Can anyone help?

best regards

Quote:

---

Originally Posted by s_ingram

Hi folks,

I read the following example in my maths book and feel the authors have not explained what they are doing very clearly. We have to find the equations of any straight lines that are invariant under the following transformation:



Where x' and y' are the image points of x and y. Any straight line will be of the form y = mx + c and any point on the line will have coordinates of the form (k, mk + c).

Thus



=



Now if y = mx + c is invariant, then this point (x', y') must also lie on y = mx + c

i.e.

giving: (equation 1).

This equation must be true for all values of k (i.e. equation 1 is an identity in k). Thus:

and

So: for those who have followed so far, I have to ask, how is equation 1 split into two parts? k is a variable and I know the difference between an equation and an identity (one is for all k, the other only particular values of k) but I don't understand the theory behind the way equation 1 suddenly becomes two equations both equal to zero. Can anyone help?

best regards

---

If Ax+ B= Cx+ D for all x then, in particular, it is true for x= 0. Setting x= 0, the equation becomes B= D. But then our equation becomes Ax+ B= Cx+ B and, subtracting B from both sides, Ax= Cx. Again, since this is true for all x, it is true for x= 1. Setting x= 1, we have A= C.

In this case, 4k - 2mk - 2c - 1 = m(-6k + 3mk + 3c + 4) + c or (4- 2m)k- (2c+1)= (-6m+ 3m^2)k+ (3mc+ 4m), A= 4-2m, B= -(2c+1), C= -6m+ 3m^2, and D= 3mc+4m. A= C gives 4- 2m,= -6m+3m^2 and B= D gives -(2c+1)= 3mc+4m.

(Strictly speaking, here they are using a variation of this: if Ax+ B= 0 for all x, then A= 0 and B= 0. Here, C= D= 0.)

More generally, if two polynomials are equal for all x then "corresponding coefficients", that is, coefficients of the same power of x, must be equal.

Dear HallsofIvy,

Thank you once again for a very clear and patient explanation! You must be a mathmatician!