4
$\begingroup$

So I am learning how to differentiate now, and I came across this problem $$f(x)=\frac{1-x}{2+x}$$ We are wanted to find $f'(x)$.

When I use $$\lim_{h\to0}\frac{f(x+h)-f(x)}{h}$$ I find that $f'(x)=\frac{-3}{(2+x)^2}$ but, when I try to find $f'(x)$ the easy way. i.e $\frac{d}{dx}x^=nx^{n-1}$. I cannot do it for some reason. Is it not possible to use that derivatives property when dealing with quotients? I know we can use it polynomial addition, subtraction and multiplication but I am struggling with quotients. Can someone please explain what it is I am not seeing?

My Attempt: $$\frac{d}{dx}\frac{1-x}{2+x}=\frac{d}{dx}(1-x)(2+x)^{-1}=(1)(-1)(2+x)^{-2}$$ Which is obviously wrong so can someone please break this down for me.:)

  • 0
    See https://en.wikipedia.org/wiki/Quotient_rule2017-02-28
  • 0
    Product rule! ${}$2017-02-28
  • 0
    How are you finding $f'(x)=\frac{-3}{(2+x)^2}$ by using the limit definition?2017-02-28
  • 0
    @StevenStadnicki if you work it out that is $f'(x)$, we did it as a class with the calculus professor.2017-02-28
  • 0
    You either want to differentiate $\dfrac{1-x}{2+x}$ using the quotient rule or you could bring the denominator "up stairs" and use the product rule on $(1-x)(2+x)^{-1}$, in this form you could also use the general power rule on the second piece.2017-02-28

4 Answers 4

4

To apply the derivative like so, you need to learn a thing called the product rule or quotient rule. To avoid it, we may proceed as follows:

$$\frac{1-x}{2+x}=\frac{3-(2+x)}{2+x}=\frac3{2+x}-1$$

The derivative of a constant is zero, hence we only need to look at

$$\frac3{2+x}=3(2+x)^{-1}\stackrel{d/dx}\to-3(2+x)^{-2}$$

However, be careful to note that

$$\frac d{dx}(1+2x)^{-1}\ne-(1+2x)^{-2}$$

Since it is not of the form $(c+x)^n$. Power rule works specifically, for now, for derivatives of the following form:

$$\frac d{dx}(c+x)^n=n(c+x)^{n-1}$$

Where $x$ is positive with a coefficient of one.

  • 0
    (+1) Thank You!:) I am going to check out the proof of this.2017-02-28
  • 1
    @NickPavini It's just the general statement that if $g(x)=\frac d{dx}f(x)$, then $g(c+x)=\frac d{dx}f(c+x)$, if you are referring to the last bit.2017-02-28
2

You can try this

$\frac{d}{dx}f(x) = \frac{d}{dx}(-1+\frac{3}{2+x}) = 3\frac{d}{dx}(2+x)^{-1} = \frac{-3}{(x+2)^2}$

1

There might be a misconception lurking here. Maybe not. In either case, I hope this helps.

Yes, by all means, the limit definition of the derivative is $$ f'(x) = \lim_{h\to0}\frac{f(x+h)-f(x)}{h}$$

Do you want a derivative of some arbitrary (well behaved) function? Say $x^2-3x$ or $\sin x$ or $e^x$? Then, sure, the above limit will give you the derivative. At least, that's true in theory. You've probably demonstrated this for polynomials in $x$, but the limit is more involved for other functions and can require a good deal of analysis.

Anyway, you next cite the "easy" method for the derivative,

$$\frac{d}{dx}x^n=nx^{n-1}$$

This isn't the derivative of any function, though. It's specifically the rule for power laws. That is, it only works for functions of the form $x^n$. It will not work, in general, for rational functions. As the others have answered, though, you might be able to use the power law with rational functions after polynomial long division.

  • 0
    Ok I see what you are saying. Thank you for that clarification and really reading my post. I definitely was under the impression that it was a general rule of sorts.2017-02-28
  • 0
    No prob. Glad it helped.2017-02-28
0

You will need to apply the quotient rule. Let the whole function be denoted $h(x)=\frac{f(x)}{g(x)}$, the numerator in your fraction be $f(x)$ and the denominator be $g(x)$. Then the derivative of your function can be found by the following forumla: $$h'(x)=\frac{g'(x)f(x)-f'(x)g(x)}{(g(x)^2}$$. Note that you can compute the derivative of $f(x)$ and $g(x)$ individually using the method you have described above, and can simply plug the results into the equation.

  • 0
    Given the level of the OP, it is extremely likely that the OP does not "need to apply the quotient rule"2017-02-28
  • 0
    When I took (high school) calculus, I learned the power rule, product rule and quotient rule all within the same week so it is definitely plausible that the quotient rule should be used.2017-02-28
  • 0
    I'm not saying you couldn't have, but by the way the question was provided, knowledge of the quotient rule seems unlikely.2017-02-28
  • 0
    You have the rule wrong as well if you look closely2017-03-01
  • 0
    I don't see anything incorrect here, feel free to point it out though.2017-03-02