I'm trying to check whether this inequality holds for large values of $B$


where $x_2>x_1$ $B>x_2$, $y_2>y_1$ and all other variables are positive. Does this inequality hold for large values of $B?$

Since the $\frac{-AB^2}{B-x_2}$ denominator is smaller than $\frac{-AB^2}{B-x_1}$ it seems intuitively to hold. However, in the limit (where $B \rightarrow \infty$, the two sides should be equal?). How can you check rigorously that this is satisfied?


Note that

$$\frac{-AB^2}{B-x_2}+\frac{2y_2}{B}<\frac{-AB^2}{B-x_1}+\frac{2y_1}{B} \iff \frac{-AB}{1-\frac{x_2}B}+\frac{2y_2}{B}<\frac{-AB}{1-\frac{x_1}B}+\frac{2y_1}{B}$$

$$ (-AB)\left(1+\frac{x_2}B\right)+\frac{2y_2}{B}<(-AB)\left(1+\frac{x_1}B\right)+\frac{2y_1}{B}$$

$$ -AB-Ax_2+\frac{2y_2}{B}<-AB-Ax_1+\frac{2y_1}{B}$$

$$\frac1B(2y_2-2y_1)< A(x_2-x_1)\iff B>\frac{2y_2-2y_1}{A(x_2-x_1)}$$

  • $\begingroup$ I think you have a mistake with the minus sign from $1-x_2/B$ to $1+x_2/B$ (from first to second line) $\endgroup$ – pafnuti Feb 8 '18 at 21:09
  • $\begingroup$ @pafnuti I've applied the binomial approximation for B large, indeed $(1+a)^{-1}\approx 1-a$ for $a\to0$ $\endgroup$ – gimusi Feb 8 '18 at 21:11

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.