Calculate linear transformation D^nS^n and S^nD^n (n= 1,2,3......) where D is differentiation and S is integration transformations. Please, help
@Kainui
Isn't this just the identity matrix?
how?
Well integrate x^2 you get \[\frac{ 1 }{ 3 } x^3 + C_1\] integrate again we get: \[\frac{ 1 }{ 12 } x^4 + C_1x+C_2\] Now take the derivative twice and you'll get back x^2. So sure this is a specific case, but it should sort of show how you could really pick anything and it would be just itself again.
if P(x) is a polynomial degree k, then, we must consider k>n, k <n , and k = n and D^n S^n \(\neq\) S^n D^n
because in the case k>n, S^n first will end at a polynomial degree (k-n) + a long tail of constants, then, take derivative back to get the original one
but if derivative first, at the n times, and k <n the polynomial = 0 , how to take the next step? for example k =2 and n =10, let say x^2 and S^10 D^10 (x^2) at D^7 the value of it disappears
Since the operator has the integral BEFORE the derivative operator you will never be losing terms. If you differentiate first, then you'd run into problems like you describe. Because the power of D is the same as S, it will always bring us back since each one exactly undoes the other amount. Unless I'm misunderstanding the problem...
OH I was only answering the first part D^nS^n and not the other part. Alright one second lol.
You'll get back a polynomial of the same degree, but depending on the power of the operator that will really decide the amount of terms get changed/added and what you get back will just be a general formula for a polynomial that you need k initial/boundary conditions to get back the exact thing you started with.
OK, that means I have to argue all cases which can happen, right? justify my answer in each case, right? even I get 0, it's a polynomial still, right?
hey, help me one more, ok?
If A and B are linear transformations such that AB - BA commutes with A, then \(A^kB-BA^k=kA^{k-1}(AB-BA)\) I stuck at k in the front from the right hand side. why is it there?
because of that k, everything goes wrong to me. :(
are you sleeping? I am sleeping, too. hehehe
Since we know it commutes, you can play around with it a little... A(AB-BA)=(AB-BA)A distribute the A's\[A^2B-ABA=ABA-BA^2\]Oh hey add them together and the ABA part cancels, that's interesting... :)\[2A(BA-BA)=(A^2B-BA^2)\]
That was a quick nap i took btw lol.
Join our real-time social learning platform and learn together with your friends!