How many differentiable functions f : R → R are there such that f ( 0 ) = 0 and
y d x d y = 2 x y ?
Note: It is intentional that y appears on both sides of the equation. Cancel at your own risk.
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Delicious Hairsplitting!
Can we change the limit x=0 to x=c where c is a constant
Log in to reply
I do not understand what you are referring to.
If you are asking "Does there exist other functions which satisfy f ( x ) = 0 for x ≤ c = 0 and f ( x ) = x 2 − c 2 otherwise, then the answer is no. The reason being that f ( x ) is not differentiable at c . (This has not been explicitly stated in the solution, which is why it is incomplete)
Log in to reply
Yes my doubt is cleared
How about f ( x ) = 0 i f x ≤ c a n d x 2 o t h e r w i s e where c > 0 ?
Superb: both problem and solution!
Problem Loading...
Note Loading...
Set Loading...
[This part explains why the answer isn't 1 or 2]
Many people would factorize the equation to obtain y ( d x d y − 2 x ) = 0 and conclude that either y = 0 or y = x 2 (using the condition that f ( 0 ) = 0 .
However, remember that the value only needs to hold pointwise , meaning that at each point, we either have y = 0 or that y = x 2 + C . In fact, there exists other solutions like:
y = { 0 x 2 x ≤ 0 x ≥ 0 , y = { x 2 0 x ≤ 0 x ≥ 0
It is clear that these 2 functions also satisfy the conditions. In particular, at y = 0 , we have d x d y = 0 whether we consider the limit from the LHS or RHS.
So why can't we mix and match 0 with x 2 + C many times? IE Why is the following solution impossible
y = ⎩ ⎪ ⎨ ⎪ ⎧ 0 x 2 − 1 0 x ≤ − 1 − 1 ≤ x ≤ 1 1 ≤ x
Looking at this specific scenario, we get that d x d y doesn't exist at x = − 1 , 1 , which is why this solution isn't valid.
[We now show that there are only 4 functions, as stated at the start.]
Let's rigourize the above observation to the general case.
First, since the function is continuous, we know that for any point a , there exists a small enough open set ( a , b ) such that ( a , b ) is identically f ( x ) = 0 or f ( x ) = x 2 + C for some C .
Second, we define a transition point t as a point such that there exists a neighbourhood ( a , b ) containing t with f ( x ) = 0 on ( a , t ) and f ( x ) = x 2 + C on ( t , b ) (or vice versa) ).
Third, at this transition point, since f is continuous at t , we must have 0 = f ( t ) = t 2 + C , or that C = − t 2 .
Fourth, at this transition point, since f is differentiable at t , calculating the left and right derivatives we get that 0 = f ′ ( t ) = 2 t .
Hence, we must have t = 0 as the only possible transition point.
As such, we have (at most) the 4 functions defined above. It is clear that these 4 functions satisfy the conditions, hence we are done.