1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Domain and Range (simple)

  1. Jan 21, 2010 #1
    1. The problem statement, all variables and given/known data

    Find the range for the function;

    fg(x) and state its range where

    [tex] f(x)= e^{2x} + 3[/tex] and [tex] g(x) = ln(x-1) [/tex]

    3. The attempt at a solution

    Putting g into f I reach;

    [tex] fg(x) = e^{2ln(x-1)}+3 [/tex] (1) - here is where I have the problem, now, I know that it simplifies down to

    [tex] fg(x) = (x-1)^2 + 3 [/tex] (2)

    However the question is, can [tex] x = 1 [/tex], it's obvious that for the function g, x > 1, or it's undefined, however when you put that into fg(x) the ln goes away so clearly in equation (2) fg >= 3, whereas with (1) fg > 3 (because x > 1). So basically I was wondering, it's obvious that g, f, & fg are linked, however is there a fundamental link between their domain/ranges? Or is it considered a new separate function?

    I'd appreciate some guidance, this was on an exam the other day.
  2. jcsd
  3. Jan 21, 2010 #2


    Staff: Mentor

    (f o g)(x) = f(g(x)) = e2ln(x - 1) + 3 = eln(x-1)2 + 3.

    Simplify the exponential term using the fact that eln(u) = u, for u > 0.

    On reading your post more carefully, I think you know this.

    The composite function f o g is tied to the two functions that make it up, but is different from both. You need to look at the domain of g, which will have a bearing on the domain of f o g. The
  4. Jan 21, 2010 #3
    Hmmm so fg(x) is not a function of x in its own right? it's dependant on the functions that make it up? That doesn't really make sense to me...=/

    As for the domain of g, I believe it's g > 0, but how does that help me?
  5. Jan 21, 2010 #4


    Staff: Mentor

    Yes f o g is a function in its own right. Note that when you write fg, it's usually interpreted to mean the product of two functions. IOW (fg)(x) = f(x)*g(x). What you're working with is the composition of two functions, which is why I have been writing the letter o between them. f o g (x) = f(g(x)).
    No that doesn't make any sense. g is the name of the function. For g(x) = ln(x - 1), the domain is {x| x > 1}. The input to the ln function has to be positive, so x - 1 > 0, so x > 1.
  6. Jan 21, 2010 #5
    Ah yeah, sorry, i did mean f(g(x)), my bad.

    So sorry for the confusion >.< but would the range therefore of f(g(x)) be >= 3? Sorry to be blunt but my teacher says it's just > 3 and i don't believe him >.<
  7. Jan 21, 2010 #6


    Staff: Mentor

    The range of f o g is {y | y > 3}. For the range to include 3, it would have to be true that eln(x - 1)2 is 0, which doesn't happen for any real number x.

    Think about it: eln(x - 1)2 is identically equal to (x - 1)2, for all x > 1. If x = 1, (x - 1)2 = 0, but eln(x - 1)2 is undefined.

    Also, x <= 1 is not in the domain of f o g, for the reason that x <= 1 is not in the domain of ln(x - 1).
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook