Newtons Inverse Roots?

Algebra Level 5

If r 1 , r 2 , r 3 , r 4 , r 5 r_1, r_2, r_3, r_4, r_5 are the complex roots of the equation x 5 3 x 4 1 = 0 x^5-3x^4-1=0 . Find the value of 1 r 1 9 + 1 r 2 9 + 1 r 3 9 + 1 r 4 9 + 1 r 5 9 . \frac{1}{r_1^9}+\frac{1}{r_2^9}+\frac{1}{r_3^9}+\frac{1}{r_4^9}+\frac{1}{r_5^9}.

Inspiration


The answer is -27.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

5 solutions

Steven Yuan
Jan 27, 2015

I will emulate the solutions posted on the linked problem, which use Vieta's.

From the given polynomial, we have

r i = 3 \sum r_i = 3

and

r i 1 = 0. \sum r_i^{-1} = 0.

The sum of the reciprocals of the roots come from the fact that the polynomial with reciprocal roots has its coefficients reversed.

x 5 3 x 4 1 = 0 x 4 ( x 3 ) = 1 x 4 = ( x 3 ) 1 x 9 = ( x 3 ) 2 x r i 9 = ( r i 3 ) 2 r i = r i 2 6 r i + 9 r i = r i 6 + 9 r i 1 = r i 6 + 9 r i 1 = 3 5 ( 6 ) + 0 = 27 . \begin{aligned} x^5 - 3x^4 - 1 &= 0 \\ x^4(x - 3) &= 1 \\ x^4 &= (x-3)^{-1} \\ x^{-9} &= \frac{(x-3)^2}{x} \\ \sum r_i^{-9} &= \sum \frac{(r_i - 3)^2}{r_i} \\ &= \sum \frac{r_i^2 - 6r_i + 9}{r_i} \\ &= \sum r_i - 6 + 9r_i^{-1} \\ &= \sum r_i - \sum 6 + 9 \sum r_i^{-1} \\ &= 3 - 5(6) + 0 \\ &= \boxed{-27}. \end{aligned}

Wow, nice solution. Didn't think that the same methods would apply here due to the differences in powers .

Trevor Arashiro - 6 years, 4 months ago

Log in to reply

Newton's sums still work, though!

Steven Yuan - 6 years, 4 months ago

Log in to reply

Yeah, just gotta reverse the coefficients

Trevor Arashiro - 6 years, 4 months ago

great solution

subharthi chowdhuri - 6 years, 4 months ago
Chew-Seong Cheong
Jan 28, 2015

It can be solved using Newton Sum method.

Let:

S 1 = i = 1 5 1 r i = 1 5 r i r j r k r l r 1 r 2 r 3 r 4 r 5 = 0 \quad \displaystyle S_1 = \sum _{i=1}^5 {\dfrac {1}{r_i}} = \sum_1 ^5 {\dfrac {r_ir_jr_kr_l}{r_1r_2r_3r_4r_5}} = 0

S 2 = 1 5 1 r i r j = 1 5 r i r j r k ( r 1 r 2 r 3 r 4 r 5 ) 2 = 0 \quad \displaystyle S_2 = \sum _1^5 {\dfrac {1}{r_ir_j}} = \sum_1 ^5 {\dfrac {r_ir_jr_k}{(r_1r_2r_3r_4r_5)^2}} = 0

S 3 = 1 5 1 r i r j r k = 1 5 r i r j ( r 1 r 2 r 3 r 4 r 5 ) 3 = 0 \quad \displaystyle S_3 = \sum _1^5 {\dfrac {1}{r_ir_jr_k}} = \sum_1 ^5 {\dfrac {r_ir_j}{(r_1r_2r_3r_4r_5)^3}} = 0

S 4 = 1 5 1 r i r j r k r l = 1 5 r i ( r 1 r 2 r 3 r 4 r 5 ) 4 = 3 1 = 3 \quad \displaystyle S_4 = \sum _1^5 {\dfrac {1}{r_ir_jr_kr_l}} = \sum_1 ^5 {\dfrac {r_i}{(r_1r_2r_3r_4r_5)^4}} = \dfrac {3}{1} = 3

S 5 = 1 r 1 r 2 r 3 r 4 r 5 = 1 1 = 1 \quad \displaystyle S_5 = \dfrac {1}{r_1r_2r_3r_4r_5} = \dfrac {1}{1} = 1

And let: P n = 1 r 1 n + 1 r 2 n + 1 r 3 n + 1 r 4 n + 1 r 5 n \quad P_n = \dfrac {1}{r_1^n} + \dfrac {1}{r_2^n} + \dfrac {1}{r_3^n} + \dfrac {1}{r_4^n} + \dfrac {1}{r_5^n}

Then:

P 9 = S 1 P 8 S 2 P 7 + S 3 P 6 S 4 P 5 + S 5 P 4 = 0 0 + 0 3 P 5 + P 4 = 3 ( S 4 P 1 + 5 S 5 ) + ( 4 S 4 ) = 3 ( S 4 S 1 + 5 S 5 ) 4 S 4 = 3 ( S 4 ( 0 ) + 5 ( 1 ) ) 4 ( 3 ) = 15 12 = 27 \begin{aligned} P_9 & = S_1P_8-S_2P_7+S_3P_6-S_4P_5+S_5P_4 \\ & = 0-0+0-3P_5+P_4 \\ & = -3(-S_4P_1+5S_5) + (-4S_4) \\ & = -3(-S_4S_1+5S_5) -4S_4 \\ & = -3(-S_4(0)+5(1)) -4(3) \\ & = -15-12 & = \boxed{-27} \end{aligned}

While the solution seems to be difficult but it can be readily calculated with a spreadsheet for any P n P_n (see below):

NICE ARRAY. Did you make that using Excel? and does S 4 S_4 denote the term in Newton's sums? Because S 4 S_4 and S 5 S_5 should be constant alone.

Trevor Arashiro - 6 years, 4 months ago

Log in to reply

Yes, it is Excel. It is easy. Yes, usually S 4 = 1 5 a i a j a k a l S_4 = \sum_1^5 {a_ia_ja_ka_l} , here it is S 4 = 1 5 1 a i a j a k a l S_4 = \sum_1^5 {\frac {1}{a_ia_ja_ka_l}} . And S 5 = 1 a 1 a 2 a 3 a 4 a 5 S_5 = \dfrac {1}{a_1a_2a_3a_4a_5} . The rest works exactly the same using Newton Sum. You can use the same array for all similar problem.

Chew-Seong Cheong - 6 years, 4 months ago
Lu Chee Ket
Jan 30, 2015

To make this simpler, we can let x = 1/ y,

We should have y^5 + 3 y - 1 = 0 and its roots shall be inverse of roots to x^5 - 3 x^4 - 1 = 0, which can save from finding inverse for sum required.

I find myself easier to solve using programmed solver rather than doing arithmetic of relating power 1 to power 9 using traditional mathematics.

0.331989029584509

0.839072433066608 + j 0.943851550132862

0.839072433066607 - j 0.943851550132862

-1.005066947858860 + j 0.937259156692892

-1.005066947858860 - j 0.937259156692892

After taking a power of 9 to each and every root,

4.89907E-05

2.077749665 + j 7.902869758

2.077749665 - j 7.902869758

-15.57777416 + j 7.939064242

-15.57777416 - j 7.939064242

Sum up to -27 + j 0 = -27

Trevor Arashiro don't you think you must correct the Language of the question? Complex roots occur in pairs!

Real numbers are also complex. For example, 5 can be written as 5 + 0 i 5 + 0i .

Steven Yuan - 6 years, 4 months ago

Log in to reply

(no offence) i said complex ROOTS. while typing i was thinking this only. But Complex roots occurs in a pair! and no denying in 5=5+0i

Agastya Chandrakant - 6 years, 4 months ago

Log in to reply

3.01 is technically a complex root. And it does have a pair. 3.01 ± 0 i 3.01\pm0i

Trevor Arashiro - 6 years, 4 months ago

Log in to reply

@Trevor Arashiro You saying of Wolfram Alpha? Haha... don't mind but sometimes it is damn wrong. I will tell you later (after posting my first question). Till then: complex root a+ib where b isn't zero. Now posting my question----->

Agastya Chandrakant - 6 years, 4 months ago

@Trevor Arashiro Solve the Integral Integration i have created! then see the answer on WolframAlpha!

Agastya Chandrakant - 6 years, 4 months ago
Trevor Arashiro
Jan 27, 2015

Hint: find r 1 9 + r 2 9 + r 3 9 + r 4 9 + r 5 9 r_1^9+r_2^9+r_3^9+r_4^9+r_5^9 of the equation x 5 + 3 x 1 x^5+3x-1

This note explains why you have to reverse the coefficients of a function to get the inverse of the roots (by inverse I mean if it has a root c, the inverse is 1 c \frac{1}{c} )

Ok, solution below


Let a = r 1 , b = r 2 . . . e = r 5 a=r_1, ~b=r_2...e=r_5

Define P n = a n + b n + c n + d n + e n P_n=a^n+b^n+c^n+d^n+e^n and

S 1 = a + b + c + d + e S_1=a+b+c+d+e

S 2 = a b + a c + a d + a e + b c + b d + b e + c d + c e + d e . . . S_2=ab+ac+ad+ae+bc+bd+be+cd+ce+de...

. .

. .

. .

S 5 = a b c d e S_5=abcde

Or more specifically, S n = S_n= the nth symmetric sum

We will be working with the equation x 5 + 3 x 1 = 0 \textbf{We will be working with the equation}~ x^5+3x-1=0

for reasons mentioned above \textbf{for reasons mentioned above}

By newtons sums, we meed to find

P 9 = P 8 S 1 P 7 S 2 + P 6 S 3 P 5 S 4 + P 4 S 5 P_9=P_8S_1-P_7S_2+P_6S_3-P_5S_4+P_4S_5

However, by vieta's, note that S 1 = S 2 = S 3 = 0 S_1=S_2=S_3=0 and S 4 = 3 S_4=3 and S 5 = 1 S_5=1 therefore (I will substitute theft values from now on), the above simplifies to

P 9 = 3 P 5 + P 4 P_9=-3P_5+P_4

We need to solve for P 5 P_5 and P 4 P_4

P 4 = 4 S 4 12 P_4=-4S_4\Rightarrow -12

(Note: the lack of the first few terms in the previous and the next is due to S 1 = S 2 = S 3 = 0 S_1=S_2=S_3=0 ).

P 5 = P 1 S 4 + 5 S 5 3 P 1 + 5 P_5=-P_1S_4+5S_5\Rightarrow -3P_1+5

Now, P 1 = a 1 + b 1 + c 1 + d 1 + e 1 P_1=a^1+b^1+c^1+d^1+e^1 and S 1 = a + b + c + d + e = 0 S_1=a+b+c+d+e=0

Therefore, P 1 = S 1 = 0 P_1=S_1=0

Plugging this value in to our equation above

P 5 = 3 P 1 + 5 = 3 ( 0 ) + 5 = 5 P_5=-3P_1+5= -3(0)+5=5

We have now established

P 5 = 5 P_5=5 , P 4 = 12 P_4=-12 , S 4 = 3 S_4=3 , S 5 = 1 S_5=1

Plugging these into our first equation.

P 9 = ( 3 ) ( 5 ) + ( 1 ) ( 12 ) = 27 P_9=-(3)(5)+(1)(-12)=\boxed{-27}

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...