1
$\begingroup$

If the roots of $$ax^2+bx+c =0 $$ lie between 1 and 2 then how can if find whether $$9a^2 +6ab +4ac$$ is positive or negative? I tried the problem by using the condition that if roots lie between a certain range $(m,n)$ then $$ af(m)f(n) >0$$ and also $m<-b/2a<n$ but that didn't helped me to determine the nature of expression $$9a^2 +6ab +4ac$$.

$\endgroup$
  • $\begingroup$ f is an unknown. $\endgroup$ – William Elliot Oct 31 '18 at 2:30
  • $\begingroup$ f() represents a general quadratic function here $\endgroup$ – Nalin Yadav Oct 31 '18 at 5:01
  • $\begingroup$ That's not the part of question.it was part of my attempt $\endgroup$ – Nalin Yadav Oct 31 '18 at 5:02
  • $\begingroup$ Sorry if it confused you $\endgroup$ – Nalin Yadav Oct 31 '18 at 5:02
  • $\begingroup$ Not defining f makes everything from there on useless. $\endgroup$ – William Elliot Oct 31 '18 at 11:12
2
$\begingroup$

This claim is false if $a=0$, as the expression is then $0$ and neither positive nor negative. I am assuming in the rest of this proof that $a \neq 0$.


Let $f(x)$ be your original function, $ax^2+bx+c$. We are given that it's roots lie in $[1, 2]$, and therefore there exist $r, s \in [1, 2]$ such that $f(x)=a(x-r)(x-s)=ax^2-2a(s+r)x+asr$. From here, we are then looking to prove that $9a^2+6ab+4ac$ is either always positive or always negative. But $$\begin{aligned} 9a^2+6ab+4ac&=9a^2-12a^2(s+r)+4a^2sr \\ &=a^2(9-12(s+r)+4sr). \end{aligned}$$ Now since $a^2$ is always positive, this is equivalent to asking whether $g(s,r) = 9-12(s+r)+4sr$ is either always positive or always negative on the range $(r,s) \in [1, 2]^2$. Luckily for us, this function is linear in $r$ and $s$, and so it suffices to check the corners: $$\begin{aligned} g(1, 1)&=-11 \\ g(1, 2)&=-19 \\ g(2, 1)&=-19 \\ g(2, 2)&=-23 \end{aligned}$$

and so $g$ is always negative

$\endgroup$
  • $\begingroup$ @ DreamConspiracy "This claim is false if a=0, as the expression is then 0 and neither positive nor negative " This the most elementary assumption we make before studying quadratic equation that leading coefficient is not zero . I didn't wrote it as I thought it was obvious $\endgroup$ – Nalin Yadav Nov 1 '18 at 4:38
  • 1
    $\begingroup$ @NalinYadav good point. I'll keep it in there for clarity anyway $\endgroup$ – DreamConspiracy Nov 1 '18 at 4:51

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.