Posted by Suzy on Sunday, November 28, 2010 at 5:30pm.
First, we have to establish that the function f(x) is invertible on the interval [0,1].
f(x) consists of a polynomial with all positive terms, so it is strictly increasing and consequently one-to-one and onto. Thus f^{-1}(x) exists.
Where x can be solved explicitly for y, an analytic expression of the inverse can be found. In other cases, we can resort to numerical solutions, which can be obtained to any accuracy we wish. For the given problem, we will supply a numerical solution.
We start with a property of f^{-1}(x). which can be looked at as
f^{-1}(f(x)) = x for all x on the given interval.
Thus f^{-1}(f(3/8))= 3/8.
To find y=f^{-1}(8/5f(3/8)), we need to find y for which
f(y)=8/5f(3/8)
then
f^{-1}(f(y))
=f^{-1}(85f(3/8))
=y
We can find the approximation by the secant method.
First evaluate
8/5f(3/8)
=2.243506202480352
Knowing f(0)=1, f(1)=32,
we give a first approximation of
y0=0.7
f(0.7)=2.115
f(0.8)=2.609
The new y could be interpolated:
y1=0.7+(0.8-0.7)*(2.2435-2.115)/(2.609-2.115)
=0.72602
We get successively
y2=0.73172
y3=0.73179
....
y=0.7317902526225
Check:
f(y)=2.243506202480352 OK
Therefore:
f^{-1}(8/5f(3/8))=0.7317902526225
IDK!!!
but the answer above needs to be put in simpler words to understand it