Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

$$K = \frac{-(\sigma^2 + 3\sigma + 2)}{\sigma^2 - 8\sigma +15}$$

How is this differentiated with respect to $\sigma$?

The answer is

$$\dfrac{dK}{d\sigma} = \frac{11\sigma^2 - 26\sigma -61}{(\sigma^2 - 8\sigma + 15)^2}$$

And the quotient rule is a very long process, I get the feeling there was a short method.

share|cite|improve this question
It's just the quotient rule. $(f/g)'={f'g-g'f\over g^2}$. Your comment about partial fractions leads me to say "the derivative, not the indefinite integral"... – David Mitra Nov 8 '11 at 12:58
That would be the quotient rule... partial fractions is an integration technique, not differentiation. – process91 Nov 8 '11 at 12:58
Er I meant quotient, I was tired – Supernovah Nov 8 '11 at 12:59
Oh thanks, I didn't know that rule – Supernovah Nov 8 '11 at 13:06

The three comments below are not really shortcuts. More like longcuts. Because $x$ is easier to type than $\sigma$, the variable has been changed. We show how to differentiate $\frac{x^2+3x+2}{x^2-8x+15}$ and leave it to someone else to change the sign at the end.

$1$) We are being asked to divide $x^2+3x+2$ by $x^2-8x+15$. So divide. We get $$\frac{x^2+3x+2}{x^2-8x+15}=1+\frac{11x-13}{x^2-8x+15}.$$ The derivative of $1$ is $0$, so we want the derivative of $$\frac{11x-13}{x^2-8x+15}.$$ The algebra of the Quotient Rule is definitely easier with a linear polynomial on top than if we use the Rule on the original expression.

$2$) If we want to practice partial fractions before they are needed for integration, we can simplify further. Note that $x^2-8x+15=(x-3)(x-5)$. The partial fractions process shows that $$\frac{11x-13}{x^2-8x+15}=-\frac{10}{x-3}+\frac{21}{x-5}.$$ Now differentiation is genuinely easy.

$3$) Logarithmic differentiation can be useful. We will be deliberately sloppy, and not worry about the fact that $\log u$ is not defined when $u \le 0$. One can show that negative $u$ in fact give no problem. The derivative of $\log|u|$ with respect to $u$ is $\frac{1}{u}$. Breaking up the interval so that we can deal with $\log$ properly gives the same derivative as the one we get if we just heedlessly calculate. Let $f(x)=\frac{x^2+3x+2}{x^2-8x+15}$. Then $$\log f(x)=\log(x^2+3x+2)-\log(x^2-8x+15).$$ Differentiate. We get more or less instantly $$\frac{f'(x)}{f(x)}=\frac{2x+3}{x^2+3x+2}-\frac{2x-8}{x^2-8x+15}.\qquad (\ast)$$ If we have to "simplify" $(\ast)$ to the form $\frac{P(x)}{Q(x)}$ where $P(x)$ and $Q(x)$ are polynomials, the simplification will not be much fun. But the expression $(\ast)$ is certainly pleasant enough if we just want to evaluate the derivative at a particular numerical value of $x$.

For complicated products/quotients, "logarithmic differentiation" can be, for certain purposes, much more efficient than conventional differentiation. However, if we want to find where the derivative vanishes, the advantage tends to evaporate.

share|cite|improve this answer

The quotient rule is not too long...

$$\begin{align*}\frac {dK}{d\sigma} =& \frac {(-2\sigma-3)(\sigma^2-8\sigma+15)-(2\sigma-8)(-\sigma^2-3\sigma-2)}{(\sigma^2-8\sigma+15)^2}\\ =& \frac{-(2\sigma^3-13\sigma^2+14\sigma+45)+(2\sigma^3-2\sigma^2-12\sigma-16)}{(\sigma^2-8\sigma+15)^2}\\ =& \frac{11\sigma^2-26\sigma-61}{(\sigma^2-8\sigma+15)^2}\end{align*}$$ The squared denominator is a good clue that the book (or whatever reference you are using for the answer) used the quotient rule.

I don't see any shortcuts to this question, so if you are being asked to differentiate it you should be familiar with the quotient rule at this point.

share|cite|improve this answer
He could use so-called ''logarithmic differentiation''; but I don't think this would qualify as a ''shortcut''. – David Mitra Nov 8 '11 at 13:26
Actually, OP could've used a combination of the product rule and chain rule if he didn't remember how to differentiate a quotient... – J. M. Nov 8 '11 at 13:35
Definitely true, although neither seems to be much of a "shortcut" (as you mentioned). Perhaps it's worth mentioning that the quotient rule is simply an application of the product rule and the chain rule? Although that's probably in the Wikipedia entry... – process91 Nov 8 '11 at 13:37

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.