PDA

View Full Version : Proving a Continuous Function has a Fixed Point


JellyFox
Oct31-10, 12:50 PM
This is a question from the exam for the calculus class I took last semester:

http://i54.tinypic.com/1z5m1zp.jpg

It looks like it might be able to be done with squeeze theorem, but I can't work it out. Please help me with this, before I descend into madness.

micromass
Oct31-10, 02:04 PM
Consider the function g(x)=f(x)-x. Does this ever cross the X-axis? (HINT: the function g is sometimes positive and sometimes negative)

JellyFox
Oct31-10, 09:57 PM
Consider the function g(x)=f(x)-x. Does this ever cross the X-axis? (HINT: the function g is sometimes positive and sometimes negative)

But it isn't necessarily negative. If f(x) = 1, then the minimum value of g is is 0, or if f(x) = 1/3 + 2x/3 then the minimum value of g is 0 again. And they show that there's a fixed point, but I don't know how to show that g has a root for all functions f. Thanks.

Landau
Nov1-10, 07:28 AM
But it isn't necessarily negative. If f(x) = 1, then the minimum value of g is is 0, or if f(x) = 1/3 + 2x/3 then the minimum value of g is 0 again. Correct, but g necessarily takes the value 0 somewhere, right? That's all you need.

(Or we could say that micromass's hint is correct, provided that 'positive' resp. 'negative' is defined as \geq 0 resp. \leq 0, instead of > 0 resp. <0 :) )

micromass
Nov1-10, 09:41 AM
Yes, I'm sorry, a function is positive for me if it is \geq 0 . Sorry for the confusion...