enter image description here

I am having little confusion regarding this graph , as we see that just before close to zero and to the right of zero the function changes its sign from negative to positive so then how come $x=0$ is a point of inflection. According to me $x=0$ should be a point of local minimum here .

  • 1
    $\begingroup$ Look again at the definition of local minimum. $\endgroup$ – Robert Israel Jul 12 '15 at 18:54
  • $\begingroup$ What makes you think it should be a minimum? (and the notes should have the singular minimum and maximum rather than minima and maxima - which are plural forms) $\endgroup$ – Mark Bennet Jul 12 '15 at 18:56
  • $\begingroup$ $0$ is greater than the neighborhood point $(x,y)$ which is in the third quadrant, so it's definitely not the local minimum. $\endgroup$ – MathGod Jul 12 '15 at 19:00
  • $\begingroup$ And is that book class 12th NCERT? $\endgroup$ – MathGod Jul 12 '15 at 19:01
  • $\begingroup$ The function $$y = \begin{cases} x^2 & x<0\\ \sqrt x & x\ge 0 \end{cases}$$ has a point of inflexion and local minimum at $x=0$. $\endgroup$ – peterwhy Jul 12 '15 at 19:02

The first derivative of the function must take different signed values at two sides of the local minimum point.

This is not satisfied in the example above. Here, $(0,0)$ is an inflection point because $f''$ changes its sign at $x=0$.

  • $\begingroup$ Is it necessary for the derivative to change sign at two sides of a local minimum? I don't think so. The derivative may oscillate that fast and change sign but still the function to obtain a local minimum or maximum. The answer to the question is no and it follows directly from the definitions. $\endgroup$ – Tolaso Jul 12 '15 at 19:22
  • $\begingroup$ Could you please me an example? $\endgroup$ – frosh Jul 12 '15 at 19:24
  • $\begingroup$ Have a look at this classic example: $$f(x)= \left\{\begin{matrix} x^2\sin \frac{1}{x} & , & x \neq 0\\ 0&, &x =0 \end{matrix}\right.$$ Both the derivative and the function oscillate near zero but still $f$ has a local minimum at $x=0$. $\endgroup$ – Tolaso Jul 12 '15 at 19:34
  • $\begingroup$ I updated my example since the function was not differentiable at $x=0$. $\endgroup$ – Tolaso Jul 12 '15 at 19:35
  • $\begingroup$ Another made up example, $$g(x) = \begin{cases} x& x<0\\ x-42 & x\ge 0 \end{cases}$$ has a local minimum at $x=0$. The function is not even continuous at $x=0$, but $g(0)$ is smaller than neighbouring $g(x)$'s, and the first derivative "at the two sides of the local minimum point" has the same sign. $\endgroup$ – peterwhy Jul 12 '15 at 19:47

Let's try to clear things up.

At the graph you are given the graph of the function $f(x)=x^3$. It is apparently differentiable on $\mathbb{R}$ as polynomial with a derivative of $f'(x)=3x^2 \geq 0$. Note that $f'(x)=0 \Leftrightarrow x=0$. One could speculate here that $x=0$ is an extrema value for $f$. No, this is not the case because the derivative does not change sign around $0$.

Hence $f$ is stricly increasing on $\mathbb{R}$. Now $f'$ is again differentiable as a polynomial. Differentiating once more we get that $f''(x)=6x$. Now, we note that $f''(x)=0 \Leftrightarrow 6x =0 \Leftrightarrow x =0$.

We also see that around zero the second derivative changes sign , meaning that $f$ has an inflection point at $x=0$ according to a known theorem

Regarding your question if a function has an extrema value at $x_0$ then at that value cannot have an inflection point and vice versa.


I leave the details up to you, but you may begin from $f''$ and study the monotony of $f$. You'll also use the process of elimination. Also you will use Fermat's theorem since $x_0$ is an internal point.

P.S 1: If for a function $f$ holds that $f'(x_0)=0$ then $x_0$ is not necessarily an extrema value of $f$. See $f(x)=x^3$.

P.S 2: Same holds if $f''(x_0)=0$ then it is not necessary an inflection point.

Hope this clears things up a bit.

  • $\begingroup$ I may be mistaken, but I don't think your proof works for $f(x)=x^4$ (since f has a local minimum at 0, but $f^{\prime\prime}(0)=0$). $\endgroup$ – user84413 Jul 12 '15 at 20:12
  • $\begingroup$ @user84413 Yes, you are right. I used the converse which does not hold necessariy. I'll try to fix the proof. I sure have one written down in my notebooks. $\endgroup$ – Tolaso Jul 12 '15 at 20:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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