MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I ask about a possible method to find the solution of algebraic equations of the form


where $a,b,c,x,y$ are real constants and $n$ is an integer. Maybe there is a simple method, but I cannot find it.

share|cite|improve this question
up vote 1 down vote accepted

If $\log_x(y) = j/k$ is rational, this reduces to a polynomial in $x^{1/k} = y^{1/j}$. Otherwise you're unlikely to get a closed form. You might use numerical methods, or a series expansion: if $y = x^r$, $$ n = \frac{\ln(-c/a)}{\ln(x)} + \sum_{k=0}^\infty \frac{(-c/a)^{kr}(b/c)^k}{k! \ln(x)} \prod_{j=1}^{k-1} (kr - j)$$

share|cite|improve this answer
Thank you very much. – China-Hong Kong Nov 19 '12 at 17:25
What about the case where $a,b,c,x,y$ are real valued functions on certain w (the unknown variable) and $n$ is a fixed integer. – China-Hong Kong Nov 19 '12 at 17:35
If you want a series solution, you'll want to expand around a known solution in powers of some parameter. Thus suppose you write your equation as $F(w) - \epsilon G(w) = 0$, where $F$ and $G$ are analytic, and $F(0) = 0$ so that $w = 0$ is a solution when $\epsilon = 0$, while $G(0) \ne 0$ and $F'(0) \ne 0$. Then the Lagrange inversion theorem gives a series expansion for $w$ in powers of $\epsilon$, convergent for sufficiently small $\epsilon$. – Robert Israel Nov 19 '12 at 19:29
Ok, Thank you very much. I will try to do this. – China-Hong Kong Nov 19 '12 at 20:14

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.