Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
0% found this document useful (0 votes)
41 views

Systems of Equations: Discussion

This document discusses techniques for solving systems of equations, including using identities, symmetric polynomials, iteration, factorization, and more. It provides examples of applying these techniques to solve problems involving systems of equations. The document is intended as a lecture on effectively combining equations.

Uploaded by

Jirayus Jinapong
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
41 views

Systems of Equations: Discussion

This document discusses techniques for solving systems of equations, including using identities, symmetric polynomials, iteration, factorization, and more. It provides examples of applying these techniques to solve problems involving systems of equations. The document is intended as a lecture on effectively combining equations.

Uploaded by

Jirayus Jinapong
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Systems of Equations

Thomas Mildorf
September 29, 2007

Solving systems of equations quickly is about exploiting symmetry.

Discussion
The canonical method for solving systems of equations is elimination by substitution. That
is, successively solving for a variable and substituting it into the other equations. As an
example, consider the system

x+y+z = 6
x + 2y + z = 8
x + y + 3z = 12.

We would begin the elimination by isolating x in the first equation as x = 6 − y − z, then


substitute x in the other equations:

(6 − y − z) + 2y + z = 6 + y = 8
(6 − y − z) + y + 3z = 6 + 2z = 12,

discovering the values y = 2 and z = 3. Backsubstituting then gives x = 6−y−z = 6−2−3 =


1, and we have the solution. This strategy is fairly reliable for simple systems of equations,
but becomes tedious for large ones. It also fails when we are required to manipulate in more
exotic ways; perhaps we cannot even solve for an isolated variable.
This lecture is about combining equations more effectively.

• Identities. It is imperative that we are comfortable with some basic factorizations.


For any complex numbers1 x and y, we have

x2 −y 2 = (x−y)(x+y), x3 −y 3 = (x−y)(x2 +xy+y 2 ), x3 +y 3 = (x+y)(x2 −xy+y 2 ).


1

That is, numbers of the form a + bi, where a and b are real numbers and i = −1 is the imaginary unit.
If you are not familiar with them, you should read about how they multiply in polar form, and how they
provide two additional (non-real) solutions to x3 = 1.

1
Example. (HMMT 2007/2) Two reals x and y are such that x−y = 4 and x3 −y 3 = 28.
Compute xy.
Answer: −3. We have 28 = x3 −y 3 = (x−y)(x2 +xy +y 2 ) = (x−y)((x−y)2 +3xy) =
4 · (16 + 3xy), from which xy = −3.

• Symmetric polynomials. Define Ik to be the set of all k-element subsets of


{1, 2, . . . , n}. For variables x1 , . . . , xn , the kth elementary symmetric polynomial is de-
fined by XY
σk := xi ,
I∈Ik i∈I

that is, the sum over all k-term products consisting of terms from {x1 , . . . , xn }. It can
be shown that if P (x1 , . . . , xn ) is any symmetric polynomial in x1 , . . . , xn , then there
exists a polynomial Q(y1 , . . . , yn ) such that

P (x1 , . . . , xn ) = Q(σ1 , . . . , σn ).

In particular, if sk denotes the sum of the kth powers of x1 , . . . , xk , then

s1 = σ1
s2 = σ12 − 2σ2
s3 = σ13 − 3σ1 σ2 + 3σ3
s4 = σ14 − 4σ12 σ2 + 2σ22 + 4σ1 σ3 − 4σ4 ,

and so on. The ability to work with symmetric polynomials is so important that we
have included it twice.
Example. (HMMT 2007/5) A convex quadrilateral is determined by the points of
intersection of the curves x4 + y 4 = 100 and xy = 4; determine its area.

Answer: 4 17. By symmetry, the quadrilateral is a rectangle having x = y and
x = −y as axes of symmetry. Let (a, b) with a > b > 0 be one of the vertices. Then
the desired area is
√  √  √ √ √
2(a − b) · 2(a + b) = 2(a2 −b2 ) = 2 a4 − 2a2 b2 + b4 = 2 100 − 2 · 42 = 4 17.

Note how we did not need to find the exact coordinates of the rectangle. In fact, we
could solve for them, but this is a waste of time. In many contests, including the AMC,
economy of computation is important in several ways. It is quicker, but it also reduces
the drain of a problem and keeps the likelihood of silly errors low.

• Iteration. Consider solving for f (a) given that f obeys some equation

f (x) + f (g(x)) = h(x), .

2
where g and h are given. It is impossible to find f (a) unless g (k) (a) = a for some
positive integer k, where g (k+1) (x) = g(g (k) (x)) (one just recovers an unbound system
of equations.) Virtually any problem of this form must be solvable by iteration of one
sort or another.
Example. (HMMT 2007/8) Let A := Q \ {0, 1} denote the set of all rationals other
than 0 and 1. A function f : A → R has the property that for all x ∈ A,
 
1
f (x) + f 1 − = log |x|.
x
Compute the value of f (2007).2
Answer: log (2007/2006) . Let g : A → A be defined by g(x) := 1 − 1/x; the key
property is that
1
g(g(g(x))) = 1 − = x.
1 − 1−1 1
x

The given equation rewrites as f (x) + f (g(x)) = log |x|. Substituting x = g(y) and x =
g(g(z)) gives the further equations f (g(y)) + f (g (g(y))) = log |g(x)| and f (g (g(z))) +
f (z) = log |g(g(x))|. Setting y and z to x and solving the system of three equations for
f (x) gives
1
f (x) = · (log |x| − log |g(x)| + log |g(g(x))|) .
2
2006 −1
For x = 2007, we have g(x) = 2007 and g(g(x)) = 2006 , so that
2006 −1
log |2007| − log 2007 + log 2006
f (2007) = = log (2007/2006) .
2
• Factorization. The fundamental theorem of algebra holds that any nonconstant
polynomial with complex coefficients has a complex root. In particular, we may write
n
Y
p(x) = xn + a1 xn−1 + · · · + an = (x − xi ), (∗)
i=1

where x1 , . . . , xn are the roots of p. Considering the distributive property, we recover


Vieta’s formula
σk = (−1)k ak ,
where σk is the kth symmetric polynomial in x1 , . . . , xn . There are many general oper-
ations on polynomials: reversing the coefficients, negating every other coefficient, etc.
These can analyzed in (*). You might try substituting to eliminate the second highest
order term, which can have the effect of forcing a subtle decomposition to become more
clear.
2
If you’ve never seen log before, you should investigate it at some point. For now, know that log x defined
to be the real exponent to which 10 may be raised to equal x (it doesn’t exist if x ≤ 0.) It has various
properties, chief among them log a + log b = log(ab).

3
Problems
1. (AMC10B 2006/12) The lines x = y/4 + a and y = x/4 + b intersect at the point (1, 2).
What is a + b?

2. (HMMT 2001) Find x − y, given that x4 = y 4 + 24, x2 + y 2 = 6, and x + y = 3.

3. (AMC10A 2006/6) What nonzero real value for x satisfies (7x)14 = (14x)7 ?

4. (AMC10A 2004/4) What is the value of x if |x − 1| = |x − 2|?

5. (AMC10A 2005/3) The equations 2x + 7 = 3 and bx − 10 = −2 have the same solution


x. What is the value of b?

6. (HMMT 2006) Find all real solutions (x, y) of the system x2 + y = 12 = y 2 + x.

7. (AMC10A 2006/8) A parabola with equation y = x2 + bx + c passes through the points


(2, 3) and (4, 3). What is c?

8. (AMC10A 2003/5) Let d and e denote the solutions of 2x2 + 3x − 5 = 0. Compute


(d − 1)(e − 1).

9. (AMC12A 2007/7) Let a, b, c, d, and e be five consecutive terms in an arithmetic se-


quence, and suppose that a + b + c + d + e = 30. Which of a, b, c, d, or e can be
found?

10. (AMC10B 2006/14) Let a and b be the roots of the equation x2 − mx + 2 = 0. Suppose
that a + (1/b) and b + (1/a) are the roots of the equation x2 − px + q = 0. What is q?

11. (HMMT 2004) Find all real solutions to x4 + (2 − x)4 = 34.


q
12. (AMC12A 2007/17) Suppose that sin a + sin b = 53 and cos a + cos b = 1. Compute
cos(a − b).

13. (AMC10A 2003/11) The sum of the two 5-digit numbers AM C10 and AM C12 is
123422. Find A + M + C.

14. (HMMT 2002) The real numbers x, y, z, w satisfy

2x + y + z + w = 1
x + 3y + z + w = 2
x + y + 4z + w = 3
x + y + z + 5z = 25.

Find the value of w.

4
15. (HMMT 2005) Find the sum of the x-coordinates of the distinct points of points of
intersection of the plane curves given by x2 = x + y + 4 and y 2 = y − 15x + 36.
√ √ √
16. (2006/6) Let a, b, c be the roots of x3 − 9x2 + 11x − 1 = 0, and let s = a + b + c.
Find s4 − 18s2 − 8s.

17. (AIME2 2003/9) The roots of x4 − x3 − x2 − 1 = 0 are a, b, c, d. Find p(a) + p(b) +


p(c) + p(d), where p(x) = x6 − x5 − x3 − x2 − x.

18. (HMMT 2001) Find the real solutions of (2x + 1)(3x + 1)(5x + 1)(30x + 1) = 10.

Homework
1. Let the polynomial P (x) = x3 + 14x2 − 30x + 15 have roots a, b, and c. Compute the
values of a + b + c, a2 + b2 + c2 , and 1/a + 1/b + 1/c.
Outline. We read off σ1 = −14, σ2 = −30, σ3 = −15. Now

a + b + c = σ1
a + b2 + c2 = σ12 − 2σ2
2

1/a + 1/b + 1/c = (ab + bc + ca)/(abc) = σ2 /σ3 .

2. Suppose x, y, and z are complex numbers such that

x+y+z = 1
x2 + y 2 + z 2 = 3
x3 + y 3 + z 3 = 7.

Find xy + yz + zx and xyz.


Outline. We are given s1 , s2 , and s3 . We seek the elementary symmetric polynomials
σ2 and σ3 . Use

s1 = σ1
s2 = σ12 − 2σ2
s3 = σ13 − 3σ1 σ2 + 3σ3

3. (HMMT 2002) Real numbers a, b, c satisfy the equations a+b+c = 26, 1/a+1/b+1/c =
28. Find the value of
a b c a c b
+ + + + + .
b c a c b a
Outline. Multiply the two given equations together and subtract three.

5
4. Let R0 denote the set of nonzero real numbers. The real valued function f is such that
 
1 1
2f (x) + f = x2 − 2
x x

for all x ∈ R0 . Find f (2).


Outline. Consider plugging in 2 and 1/2. One recovers a system of two equations
and two unknowns.

5. Solve for all real x such that



3

3
38 − x + 38 + x = 4.

3

3
Outline. Set a = 38 − x and b = 38 + x. Then a + b = 4 and

76 = a3 + b3 = (a + b)(a2 − ab + b2 ) = 4 · (16 − 3ab),

so that ab = −1. Then solve the quadratic.

You might also like