2
$\begingroup$

Take an example polynomial $f(x, y) = y^2 x + y^3 - x^2$. A solution to $f(x,y)=0$ exists with Puiseux series given by $y(x) = x^{2/3} - x/3 + x^{4/3}/9+\cdots$. I got this by having Mathematica directly solve $f=0$ for $y$ and then perform a Puiseux series expansion for me about $x=0$.

However, since a direct solution to $f=0$ is obviously not available for arbitrary $f$, what I'd like is a method to determine just the lowest-order term of the expansion of $y(x)$ about $x=0$ for arbitrary $f(x,y)$. In the example above, this term would be $x^{2/3}$.

Thanks!

$\endgroup$

2 Answers 2

4
$\begingroup$

Finding such solutions was in fact Newton's original motivation for inventing "Newton polygons". There's a nice exposition of this in Chapter 2 of "The Implicit Function Theorem: History, Theory, and Applications" by Steve Krantz and Hal Parks.

$\endgroup$
2
$\begingroup$

The magic words are "Newton Polygon". (there is a vast literature on the subject, the wikipedia article is one place to start).

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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