[Math] Finding local max/min of a function

calculus

I am having difficulty finding the local maxima and minima on the function: $$f(x)=\frac{x^2}{x-a}$$ on the invterval $(0, \infty)$.

so far I have worked out $$f'(x)=\frac{x(x-2a)}{(x-a)^2}$$ using the quotient rule.

I have also found the critical points $x=2a$ and $x=a$

I however am stuck after this point. I have tried to use the first derivative test to find out what values are local mins and local maxs, but I haven't attempted a problem where x isn't a set value. While I understand a is a constant it isn't a set value and this confuses me as to how to use it in the first derivative test.

Note: It mentions the answer may invovle a parameter of a which makes me think the answer is dependent on a?

Best Answer

Note that $a=0$ is special. Apart from not being defined at $0$, our function is then the same as the function $g(x)=x$, and has no local max or min.

We deal now with $a\gt 0$ and $a\lt 0$. I would deal with these separately.

Case $a\gt 0$: The derivative is $0$ at $x=0$ and at $x=2a$. We examine each of these points in turn.

(i) The bottom of the derivative is safely positive. Consider the top $x(x-2a)$ very near $x=2a$. Note that $x$ is positive. For $x$ near $2a$ but smaller than $2a$, we see that $x-2a$ is negative. For $x\gt 2a$, it is positive. so the derivative is negative for $x$ smaller than $2a$, but near $2a$, and positive afterwards. Thus our function near $2a$ is going down and then up, we have a local minimum at $x=2a$.

(ii) We now look at the derivative near $x=0$. If $x\lt 0$, then $x$ is negative. Also, $x-2a$ is negative there, so the derivative is positive. For $x\gt 0$ but close to $0$, we have $x$ positive and $x-2a$ negative, so the derivative is negative. Thus near $x=0$ our function is going up and then down, we have a local max at $x=0$. But the problem only asks about the interval $(0,\infty)$, so we needn't have bothered.

Case $a\lt 0$: We are only asked for local max and min in $(0,\infty)$. In the case $a\lt 0$ there are no critical points in $(0,\infty)$, so no local max or min. without that restriction, the analysis would go along lines similar to the case $a\gt 0$.