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!

Finding the range of rational functions algebraically

  1. Oct 11, 2012 #1
    how to find the range of rational functions like f(x) = [itex]\frac{1}{{x}^{2}-4}[/itex] algebraically , i graphed it and seen that (-1/4,0] can not be in range . generally i am interested in how to find the range of functions and rational functions in particular
  2. jcsd
  3. Oct 12, 2012 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    How do you think it might relate to the maxima and minima of the function? Where are those for your example?
  4. Oct 12, 2012 #3

    yes i got those points by differentiating but can't we find it without using derivatives ? i mean any algebraic method ?
  5. Oct 12, 2012 #4


    User Avatar
    Homework Helper

    I can't think of any algebraic methods, and if I were to use logical reasoning, it would still involve some kind of crude link to limits and derivatives.
  6. Oct 12, 2012 #5


    User Avatar
    Science Advisor

    By inspection |x| = 2 is critical and x = 0 is a local minimum.
  7. Oct 12, 2012 #6
    minimum ? it's max. there is no minmimum for this function .but how can i find the range though ? i mean how it's useful to find out the range exactly ?
    Last edited: Oct 13, 2012
  8. Oct 13, 2012 #7


    User Avatar
    Science Advisor

    You're right. It is local max (-1/4). As |x| -> 2 from below, f(x) -> -∞. However |x| -> 2 from above, f -> +∞. Finally as |x| -> ∞, f -> 0.

    Net result: range has two parts (-∞,-1/4) for |x| < 2, and (0,+∞) for |x| > 2.
  9. Oct 13, 2012 #8


    User Avatar
    Homework Helper

    I took a little while to think about it, and yes, there is an algebraic solution to the problem.

    For the function


    let the range be denoted R, which is also the y-value, so we have


    Now, we want to solve for x:




    Now, x exists (and thus a correspondent range exist) whenever

    [tex]\frac{1+4R}{R}\geq 0[/tex]

    and clearly the opposite of that is, if

    [tex]\frac{1+4R}{R}< 0[/tex]

    then the domain does not exist, thus the range does not exist.

    Now, if we solve this inequality, for R>0 (we can see that [itex]R\neq 0[/itex])



    And we obviously can't have that both R>0 and [itex]R<\frac{-1}{4}[/itex] so we scrap that. Now if R<0



    Which gives us the intersection [tex]\frac{-1}{4}<R<0[/tex] as required.
  10. Oct 14, 2012 #9


    User Avatar
    Science Advisor

    There are no values of x where -1/4 < y < 0. As I have indicated, the range is everything but this interval.
  11. Oct 14, 2012 #10


    User Avatar
    Homework Helper

    In my solution I wrote

  12. Oct 15, 2013 #11

    can u plz explain to me how did u reach the conclusion that R will be greater than zero? solving the inequality only yields R>-1/4
  13. Oct 15, 2013 #12


    User Avatar
    Homework Helper

    I had written that precisely a year ago :biggrin:

    Since [itex]-1/4 < R < 0[/itex] is where the range of the function does not exist, everything else is where the range does exist.

    So the range is
    [itex]y> 0 \cup y\leq -1/4[/itex]

    (edit: [itex]y\neq 0[/itex] because we can't divide by 0)

    If there's something in my earlier solution that you don't understand, just point it out.
    Last edited: Oct 15, 2013
  14. Oct 15, 2013 #13
    lol well i'm glad u're still alive :P, yes there is something i do not get, the range of that function would be the domain of the inverse function correct? 1+4R/R > or equal to 0 will give us all the possible range correct?

    solving this only gives us range bigger than -1/4, and range does not equal to 0, but as u already know this function has range above than zero, the question is how did u deduce that from the problem/inequality?
  15. Oct 15, 2013 #14


    User Avatar
    Homework Helper

    Yes and yes.

    Oh you're right about excluding R=0 from the solution, I missed that. You're not solving the inequality correctly though.

    Remember that when you multiply through by R, if R<0 then you need to change the reverse the sign of the inequality

    So for R>0
    [itex]1+4R\geq 0[/itex]

    but for R<0
    [itex]1+4R\leq 0[/itex]

    Now when you solve these two inequalities, you only keep the intersections that makes sense. For example, if we assume R<0 but then solve the inequality and find that R>1, there are no solutions for R such that both of these can hold true at the same time (we say that the intersection of the sets is the empty set). If however we assume R<0 and solve the inequality to find R>-1, then our solution set is -1<R<0.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook