1
$\begingroup$

I have the following equation and I was wondering if I can solve for x given that it appears both as an exponent and a base:

$[\frac{1}{\sqrt {2\pi}.S}.e^{-\frac{(x-M)^2}{2S^2}}-0.5\frac{1}{\sqrt {2\pi}.S}.e^{-\frac{(x-M)^2}{2S^2}}.\frac{4.(\frac{x-N}{\sqrt {2}.D})}{\sqrt {\pi}.e^{-\frac{(x-N)^2}{2D^2}}+\sqrt {\pi.e^{-2\frac{(x-N)^2}{2D^2}}+16.\frac{(x-N)^2}{2D^2}}} ]+[ \frac{1}{\sqrt {2\pi}.D}.e^{-\frac{(x-N)^2}{2D^2}}-0.5\frac{1}{\sqrt {2\pi}.D}.e^{-\frac{(x-N)^2}{2D^2}}.\frac{4.(\frac{x-M}{\sqrt {2}.S})}{\sqrt {\pi}.e^{-\frac{(x-M)^2}{2S^2}}+\sqrt {\pi.e^{-2\frac{(x-M)^2}{2S^2}}+16.\frac{(x-M)^2}{2S^2}}}] = \frac{1}{v}$

In other words, can I find an expression for x as a function of all the other variables?

$\endgroup$
  • $\begingroup$ Most likely not. $\endgroup$ – 1-___- Aug 19 '15 at 19:03
  • 2
    $\begingroup$ Wow! Your equation is very complicated! How did you get it, and why do you want to solve it? $\endgroup$ – Crostul Aug 19 '15 at 19:04
  • $\begingroup$ It's an approximation of the derivative of a joint probability function that I would like to equate to the value $\frac{1}{v}$ through x $\endgroup$ – tvbc Aug 19 '15 at 19:07
1
$\begingroup$

Let's simplify the equation a bit so we can better see its form:
$$C_{1}e^{X_{1}} + C_{2}e^{X_{1}}\cdot \frac{X_{2}}{C_{3}e^{X_{3}} \sqrt{C_{4}e^{X_{3}} + C_{5}X_{3}}} + C_{6}e^{X_{4}} + C_{7}e^{X_{4}}\cdot \frac{X_{5}}{C_{8}e^{X_{6}} \sqrt{C_{9}e^{X_{6}} + C_{10}X_{6}}} = C_{11}$$ With constants as $C_{n}$ above and expressions involving x as $X_{n}$ above.
Even if all of the constants were 1, we would get:
$$e^{X_1} + e^{X_4} + X_{2}\frac{e^{X_1}e^{-X_{3}}}{\sqrt{e^{X_{3}} + X_{3}}} + X_{5}\frac{e^{X_4}e^{-X_{6}}}{\sqrt{e^{X_{6}} + X_{6}}} = 1$$ Which could, with some further magical simplifications (letting all of the expressions in x be the same), be written as
$$e^{2X} + \frac{2X}{\sqrt{e^{X} + X}} = 1$$
Heck, even if the denominator was 1, we would get (letting $Y = 2X$):
$$e^{Y} + Y = 1$$
Which in and of itself is not solvable for Y.

All of this was just to show that even with certain simplifications your equation reduces to a form which is very much unsolvable, so it is extremely unlikely that your original equation is solvable for x.

$\endgroup$
  • $\begingroup$ Good enough for me, thanks. $\endgroup$ – tvbc Aug 20 '15 at 20:45

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.