Polynomials question (1 Viewer)

unmentionable

Member
Joined
Dec 13, 2003
Messages
112
Location
syd
Gender
Undisclosed
HSC
2004
i've been trying to do this question for ages.. its so annoying :mad:

the equation x^4 - px^3 + qx^2 - rx +s = 0 has roots alpha, beta, gamma and delta

show that if alpha x beta = gamma x delta, then r^2 = p^2s

thanks
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
ok conc on r and p and s.. since q isnt req.
alpha=a
beta = b
gamma=y
delta = d

polynomial rule
p = -(alpha + beta + gamma + delta) =-(a+b+y+d)
r = -(aby + byd + yda + dab)
s = abyd

now let ab = yd ("if alpha x beta = gamma x delta")

r^2 = (aby + byd + yda + dab)^2
= (aby + byd)^2 +2(aby + byd)( yda + dab)+ (yda + dab)^2

using the IF thing

=(yd)^2(y+b)^2+2(yd)^2(b+y)(a+d) +(yd)^2(a+d)^2
=(yd)^2 [ (y+b)^2 + 2(b+y)(a+d)+(a+d)^2]
=(yd)^2[ (y+b+a+d)^2]
=(yd)^2[ (y+b+a+d)^2]

s=abyd=(yd)^2 from the ab=yd
and p =a+b+y+d

therefore r^2 = p^2s
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
just do it systematically..
a question lik this in Q5 last year seemed hard but just think bout the basic steps..
no need to expand it all cos that will make factoring harder.. just think goal is p^2s
not a massive expansion.. massive expansion = bit*h of a factorisation
 

CrashOveride

Active Member
Joined
Feb 18, 2004
Messages
1,488
Location
Havana
Gender
Undisclosed
HSC
2006
or maybe quicker:

abcd = s
a^2.b^2 =s

SIGMA (a) = p
sigma (abc): ab( sigma (a)) = r
little re-arranging, squaring and r^2 = sp^2
 

CrashOveride

Active Member
Joined
Feb 18, 2004
Messages
1,488
Location
Havana
Gender
Undisclosed
HSC
2006
if a , b and c are roots of x<sup>3</sup> + qx + r = 0, find the cubic equation in y, in terms of q and r, whose roots are:

(b+c-2a), (c+a-2b) and (a+b-2c)

Now for these type of questions eg. find eqn whose roots are the squares i would just let y = x<sup>2</sup> and go from there. But what do i do from here, the solution just uses the first root as y and subs back in and thats it.
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,644
Gender
Male
HSC
N/A
CrashOveride said:
if a , b and c are roots of x<sup>3</sup> + qx + r = 0, find the cubic equation in y, in terms of q and r, whose roots are:

(b+c-2a), (c+a-2b) and (a+b-2c)
Let y = a + b + c - 3x.

It follows that as x takes the values a, b and c, the roots of the equation in y will be (respectively) b + c - 2a,
c + a - 2b and a + b - 2c.

Now, a + b + c = sum of roots = 0, and so we have y = -3x, or x = -y / 3

(-y / 3)<sup>3</sup> + q(-y / 3) + r = 0
On multiplication by -27, we get: y<sup>3</sup> + 9qy - 27r = 0, which is the required equation.
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
CrashOveride said:
or maybe quicker:

abcd = s
a^2.b^2 =s

SIGMA (a) = p
sigma (abc): ab( sigma (a)) = r
little re-arranging, squaring and r^2 = sp^2
how much quicker with sigma form that u propose?
 

unmentionable

Member
Joined
Dec 13, 2003
Messages
112
Location
syd
Gender
Undisclosed
HSC
2004
i did the question using pointdexter's method and it only took me a few minutes
i dont understand the sigma notation thing sorry =\
 

schmeichung

Member
Joined
Dec 19, 2003
Messages
178
Gender
Male
HSC
2004
Sorry I dont get this:
>>>Let y = a + b + c - 3x.

How would you think of this?



CM_Tutor said:
Let y = a + b + c - 3x.

It follows that as x takes the values a, b and c, the roots of the equation in y will be (respectively) b + c - 2a,
c + a - 2b and a + b - 2c.

Now, a + b + c = sum of roots = 0, and so we have y = -3x, or x = -y / 3

(-y / 3)<sup>3</sup> + q(-y / 3) + r = 0
On multiplication by -27, we get: y<sup>3</sup> + 9qy - 27r = 0, which is the required equation.
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
he got y is the way to generate the new equations roots.. from the old one ,x.
it took me a while too but im so hungova..

umm lik if i wanted the roots on new eq to be 3a,3b,3c
id let y=3x :)
 

CrashOveride

Active Member
Joined
Feb 18, 2004
Messages
1,488
Location
Havana
Gender
Undisclosed
HSC
2006
damn shud have got that, thnx CM

btw, if we have some polynomial P(x) with a zero of multiplicity r, we can express it as P(x) = (x - A)^r . Q(x) ..... where A is the zero. Why do we say also "where Q(A) != 0"
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
well if Q(A) = 0 then the multiplicity would be >r
i.e. P(x) = (x - A)^r . Q(x)
and Q(A) =0 then A is a factor of Q(x)

i.e.Q(x)= (x-A).M(x) and then P(x)=(x - A)^{r+1} . M(x)
so the multiplicity would not be r.. hope that makes sense
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,644
Gender
Male
HSC
N/A
schmeichung said:
Sorry I dont get this:
>>>Let y = a + b + c - 3x.

How would you think of this?
Most substitutions are fairly obvious.

ie. If you have roots at x = a, b and c, and you want roots at y = a + 1, b + 1 and c + 1, then let y = x + 1

If you have roots at x = a, b and c, and you want roots at y = 5a, 5b and 5c, then let y = 5x

If you have roots at x = a, b and c, and you want roots at y = a<sup>2</sup>, b<sup>2</sup> and c<sup>2</sup>, then let y = x<sup>2</sup>

However, some substitutions are more complicated. If the new roots are in terms of more than one existing root, like in the question above, then the trick is that the substitution must be symmetric in the existing roots. That is, you must be able to swap the roots a for b, b for c and c for a (for example) in the substitution expression, whilst the substitution itself remains the same. For example, y = a + b + c - x is symmetric in a, b and c, but y = a + b - c - x is not. Also, if possible, the substiutions should make use of a + b + c, ab + bc + ca or abc, as these are known from the original equation.

In thie question here, I need the roots added together, so I started with y = a + b + c + f(x), where f(x) must be a function of x such that y takes the required values as x takes the roots a, b and c. Now, when x = a, I have
a + b + c + f(a), which needs to be one of b + c - 2a, c + a - 2b or a + b - 2c. Clearly, the only possibility is
a + b + c + f(a) = b + c - 2a, as otherwise f(a) depends on b or c, which isn't possible. So, f(a) = -3a, and the required substitution is y = a + b + c - 3x

Some more examples:

Required Roots: a + b, b + c and c + a. Required Substitution: y = a + b + c - x
Required Roots: 2a + b + c, a + 2b + c and a + 2b + c. Required Substitution: y = a + b + c + x
Required Roots: a<sup>2</sup>bc, ab<sup>2</sup>c and abc<sup>2</sup>. Required Substitution: y = abcx
Required Roots: ab, bc and ca. Required Substitution: y = abc / x
Required Roots: ab / c, bc / a and ca / b. Required Substitution: y = abc / x<sup>2</sup>
Required Roots: a<sup>2</sup> + b<sup>2</sup>, b<sup>2</sup> + c<sup>2</sup> and c<sup>2</sup> + a<sup>2</sup>. Required Substitution: y = a<sup>2</sup> + b<sup>2</sup> + c<sup>2</sup> - x<sup>2</sup>
which can be more usefully written as y = (a + b + c)<sup>2</sup> - 2(ab + bc + ca) - x<sup>2</sup>

Does this make it clearer?
 
Last edited:

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
thats a real good method/ explanation!? i mostly got the obvious ones thru past experiences or it would just pop into my head.. (dun ask.. many eureka moments in 4U last year :p)..
method is quite systematic too ^^
key to a good tutor :)
 

schmeichung

Member
Joined
Dec 19, 2003
Messages
178
Gender
Male
HSC
2004
I got it now thanks a lot!

CM_Tutor said:
Most substitutions are fairly obvious.

ie. If you have roots at x = a, b and c, and you want roots at y = a + 1, b + 1 and c + 1, then let y = x + 1

If you have roots at x = a, b and c, and you want roots at y = 5a, 5b and 5c, then let y = 5x

If you have roots at x = a, b and c, and you want roots at y = a<sup>2</sup>, b<sup>2</sup> and c<sup>2</sup>, then let y = x<sup>2</sup>

However, some substitutions are more complicated. If the new roots are in terms of more than one existing root, like in the question above, then the trick is that the substitution must be symmetric in the existing roots. That is, you must be able to swap the roots a for b, b for c and c for a (for example) in the substitution expression, whilst the substitution itself remains the same. For example, y = a + b + c - x is symmetric in a, b and c, but y = a + b - c - x is not. Also, if possible, the substiutions should make use of a + b + c, ab + bc + ca or abc, as these are known from the original equation.

In thie question here, I need the roots added together, so I started with y = a + b + c + f(x), where f(x) must be a function of x such that y takes the required values as x takes the roots a, b and c. Now, when x = a, I have
a + b + c + f(a), which needs to be one of b + c - 2a, c + a - 2b or a + b - 2c. Clearly, the only possibility is
a + b + c + f(a) = b + c - 2a, as otherwise f(a) depends on b or c, which isn't possible. So, f(a) = -3a, and the required substitution is y = a + b + c - 3x

Some more examples:

Required Roots: a + b, b + c and c + a. Required Substitution: y = a + b + c - x
Required Roots: 2a + b + c, a + 2b + c and a + 2b + c. Required Substitution: y = a + b + c + x
Required Roots: a<sup>2</sup>bc, ab<sup>2</sup>c and abc<sup>2</sup>. Required Substitution: y = abcx
Required Roots: ab, bc and ca. Required Substitution: y = abc / x
Required Roots: ab / c, bc / a and ca / b. Required Substitution: y = abc / x<sup>2</sup>
Required Roots: a<sup>2</sup> + b<sup>2</sup>, b<sup>2</sup> + c<sup>2</sup> and c<sup>2</sup> + a<sup>2</sup>. Required Substitution: y = a<sup>2</sup> + b<sup>2</sup> + c<sup>2</sup> - x<sup>2</sup>
which can be more usefully written as y = (a + b + c)<sup>2</sup> - 2(ab + bc + ca) - x<sup>2</sup>

Does this make it clearer?
 

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top