0

$$ \tan^{-1}x+\tan^{-1}y=\begin{cases} \pi+\tan^{-1}\frac{x+y}{1-xy} &\mbox{if } xy>1 \quad\&\quad x,y>0\\ -\pi+\tan^{-1}\frac{x+y}{1-xy} & \mbox{if } xy>1\quad\&\quad x,y<0\end{cases} $$

The above expression can be proved, pls check. asnswer of "lab bhattacharjee" in the post A question about the arctangent addition formula.

What if I prove the conditions as follows

Attempt:

Taking $$ \alpha=\tan^{-1}x\implies x=\tan\alpha\quad\text{ ,where }\frac{-\pi}{2}<\alpha<\frac{\pi}{2}\\ \beta=\tan^{-1}y\implies y=\tan\beta\quad\text{ ,where }\frac{-\pi}{2}<\beta<\frac{\pi}{2}\\ \implies-\pi<\alpha+\beta<\pi $$ Now consider the case where $\frac{\pi}{2}<\alpha+\beta<{\pi}$ (or) -$\pi<\alpha+\beta<\frac{-\pi}{2}$ $$ \frac{-\pi}{2}<\alpha,\beta<\frac{\pi}{2}\\ \implies\color{red}{\cos(\alpha+\beta)<0},\quad\text{ as }\alpha+\beta\in[\tfrac{\pi}{2},\pi]\cup[-\pi,\tfrac{-\pi}{2}]\\ \implies\cos\alpha.\cos\beta-\sin\alpha.\sin\beta<0\\ \implies \cos\alpha.\cos\beta-\cos\alpha\tan\alpha.\cos\beta\tan\beta<0 \\\implies \cos\alpha\cos\beta\big[1-xy\big]>0\implies1-xy<0\implies \boxed{xy>1} $$ For case 1:

If $x,y<0$ then $\tan^{-1}x,\tan^{-1}y<0$

$$ \implies \tan^{-1}x+\tan^{-1}y<0 $$ which is not the case as $\frac{\pi}{2}<\tan^{-1}x+\tan^{-1}y<{\pi}$. $$ \implies \boxed{x,y>0} $$

For case 2:

If $x,y>0$ then $\tan^{-1}x,\tan^{-1}y>0$

$$ \implies \tan^{-1}x+\tan^{-1}y>0 $$ which is not the case as ${-\pi}<\tan^{-1}x+\tan^{-1}y<\frac{-\pi}{2}$. $$ \implies \boxed{x,y<0} $$

Is there anything wrong in taking $\cos(\alpha+\beta)<0$ in the case where $\frac{\pi}{2}<\alpha+\beta<{\pi}$ (or) $-\pi<\alpha+\beta<\frac{-\pi}{2}\\$ and proving as above ?

Sooraj S
  • 7,573
  • https://math.stackexchange.com/questions/2604703/derive-the-conditions-xy1-for-tan-1x-tan-1y-tan-1-fracxy1-xy?rq=1 – Guy Fsone Jan 27 '18 at 12:29
  • https://math.stackexchange.com/questions/1837410/inverse-trigonometric-function-identity-doubt-tan-1x-tan-1y-pi-tan – Guy Fsone Jan 27 '18 at 12:31
  • @GuyFsone it is proved at https://math.stackexchange.com/q/1837410/223599. My question is whether my approach is right ? – Sooraj S Jan 27 '18 at 12:34
  • See https://math.stackexchange.com/questions/326334/a-question-about-the-arctangent-addition-formula/326538#326538 – lab bhattacharjee Jan 27 '18 at 13:16
  • @labbhattacharjee that is the post i have referred in my question. My doubt is whether taking $\cos(\tan^{-1}x+\tan^{-1}y)<0$ ok in my attempt to derive the conditions $xy>1$ in case 2 and 3 ? – Sooraj S Jan 27 '18 at 13:22
  • @labbhattacharjee pls check i have edited my original post. Hope my doubt is well explained now. – Sooraj S Jan 27 '18 at 14:42
  • @GuyFsone i have edited my original post. pls check. – Sooraj S Jan 27 '18 at 14:42
  • @ss1729 I saw it. but it is too long for me to check everything now – Guy Fsone Jan 27 '18 at 14:46
  • @GuyFsone i am basically asking about the condition $\cos(\alpha+\beta)<0$ that i have chosen to reach the conditions $xy>0$ – Sooraj S Jan 27 '18 at 16:15

0 Answers0