Uncovering the Secrets of the General Quadratic Equation: A Geometric Approach

Click For Summary

Discussion Overview

The discussion centers on the general quadratic equation, represented as Ax² + Bxy + Cy² + Dx + Ey + F = 0, exploring its derivation, the role of the discriminant (B² - 4AC) in classifying conic sections, and the geometric interpretations involved in manipulating the equation.

Discussion Character

  • Exploratory
  • Technical explanation
  • Conceptual clarification
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • Some participants discuss the derivation of the general quadratic equation and its components, noting that it is a standard form rather than something that is derived.
  • There are various methods proposed for determining the discriminant, including coordinate rotation to eliminate the xy term and using eigenvalues of a matrix representation of the quadratic part.
  • One participant suggests that the discriminant can classify conics: B² < 4AC indicates an ellipse, B² = 4AC indicates a parabola, and B² > 4AC indicates a hyperbola, although another participant expresses uncertainty about how the discriminant determines the class of conic sections.
  • Some participants express confusion regarding the relationship between the general quadratic equation and simpler quadratic forms, particularly the mixing of x and y terms.
  • There is mention of degenerate conics and their characteristics, with examples provided to illustrate these cases.
  • One participant suggests using the Distance Formula to derive conic sections and discusses the implications of switching x and y in the context of graph rotation.

Areas of Agreement / Disagreement

Participants express a range of views on the derivation and classification of conic sections, with no consensus reached on the exact implications of the discriminant or the relationship between different forms of quadratic equations.

Contextual Notes

Some participants note limitations in their understanding of the geometric implications of the quadratic equation and the precise justification for the relationships between terms, indicating that further exploration may be necessary.

DJ24
Messages
21
Reaction score
0
The general quadratic equation:

Ax2 + Bxy + Cy2 + Dx + Ey + F = 0

How is this derived? How does the discriminant, B2 – 4AC determine the class of the conic section?
 
Mathematics news on Phys.org
Are you asking how B2-4AC is derived? Because the actual polynomial equation isn't derived, or something that you prove, it just is.
 
That equation, Ax^2+ Bxy+ Cy^2+ Dx+ Ey+ F= 0 is the "general quadratic equation" because you cannot have any other terms in a polynomial without introducing higher powers- those are all the terms of degree two or less. As for determining the "discriminant", there are several ways to do that, essentially "rotating" the coordinate system to get rid of the "xy" term. If you let x= x&#039; cos(\thet)- y&#039; sin(\theta) and y= x&#039;sin(\theta)+ y&#039;cos(\theta) and substitute those into the equation, you get a very complicated equation in x' and y' with coefficients that depend on \theta. Choose theta so that the coefficient of x'y' is 0 and see what kind of conic you have. You really only need to look at the "quadratic" part- you can ignore D, E, and F.

A simpler calculation, but more "sophisticated", is to write the quadratic part of the equation as a "bilinear tranformation":
\begin{bmatrix} x &amp; y\end{bmatrix}\begin{bmatrix} A &amp; \frac{B}{2} \\ \frac{B}{2} &amp; C \end{bmatrix}\begin{bmatrix} x \\ y\end{bmatrix}
and look at the eigenvalues and eigenvectors of that 2 by 2 matrix. It will be a circle the two eigenvalues are the same, an ellipse if they are different but of the same sign, a parabola if one of them is 0, and an hyperbola if they are of different signs.
 
What do you mean exactly by "rotating" the graph? Are you referring to an xy or xyz plane graph?
 
DJ24 said:
What do you mean exactly by "rotating" the graph? Are you referring to an xy or xyz plane graph?
Your equation only had x and y so, of course, I was referring to an xy graph. And by "rotating" I mean moving the axes so that the origin remains fixed while the two axes move through an angle \theta. For example, rotationg by \pi/4 radians, the x-axis moves to the line y= x and the y-axis to the line y= -x.
 
The product xy from Ax^2 + Bxy + Cy^2 + Dx + Ey + F = 0 will make the conic oblique to the coordinate axes. The substitution x = x&#039; \cos(\theta) - y&#039; \sin(\theta) and y = x&#039; \sin(\theta)+ y&#039; \cos(\theta) allow us to make the conic parallel or perpendicular to one of the coordinate axes. Talking about the discriminant B2 - 4AC, if B^2 &lt; 4AC the conic is an ellipse, if B^2 = 4AC conic is parabola, and if B^2 &gt; 4AC the conic is hyperbola. On how it exactly determine the class of conic section, I don't have an answer for now. But one thing I am certain that the discriminant defines the conic similar to how the eccentricity defines it., ie e < 1 (ellipse), e = 1 (parabola), and e < 1 (hyperbola).
 
Last edited:
The best way to see that is the "eigenvalue" method I mentioned.
The determinant of the matrix
\begin{bmatrix}A &amp; \frac{B}{2} \\ \frac{B}{2} &amp; C\end{bmatrix}
is AC- B2/4= (1/4)(4AC- B) and will be positive if B- 4AC< 0, negative if B- 4AC> 0, and 0 if B- 4AC= 0. The product of the eigenvalues is equal to the derivative so we have that the two eigenvalues are of the same sign (a circle or ellipse) if B2- 4AC< 0, of opposite sign (a hyperbola) if B2- 4AC> 0, and at least one 0 (a parabola) if B2- 4AC= 0.

(Two be complete, you should also include "degenerate conics": if A= 1, C= -1, B= D= E= F= 0, the equation is x2- y2= (x- y)(x+ y)= 0 which graph is the two lines x+ y= 0 and x- y= 0 (and is a "degenerate" hyperbola). If A= 1, B= -2, C= 1, D= E= F= 0, the equation is x2- 2xy+ y2= (x- y)2= 0, a single straight line (a "degenerate" parabola). If A= 1, C= 1, B= D= E= F= 0, the equation if x2+ y2= 0, the single point (0, 0) (a "degenerate" circle).)
 
I wasn't sure. Is it true, for the quadratic equation, that ax2 + bx + c = 0 = f(x)? I was just thinking that Ax2 + Bxy + Cy2 + Dx + Ey + F = 0 = f(x,y). I don't know; I may be missing something. The graph of a quadratic equation is by the function f(x) = ax2 + bx + c. I just don't understand what everything implies and why y-values are suddenly intermingled with x-values into the general quadratic equation.
I would also like to understand why B2 = 4AC
 
You can achieve enough justification for yourself using the Distance Formula to derive the equations for all of the conic sections according to their Distance Formula definitions. You should at least intuitively understand that switching the roles of x and y will give you graph results tilted by a Right angle. You should now see that each of the terms in the Quadratic general equation are accounted. The XY term occurs because of rotation of the graph. Justifying this precisely is beyond my own understanding, but it would involve Trigonometry. For an improved understanding of Y and Y2 terms, when the X terms are also present (or even when they are not), you can complete the square for them in order to help obtain a standard form equation or function.
 

Similar threads

  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 16 ·
Replies
16
Views
5K
  • · Replies 14 ·
Replies
14
Views
6K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 18 ·
Replies
18
Views
3K
  • · Replies 3 ·
Replies
3
Views
5K
  • · Replies 3 ·
Replies
3
Views
5K