0% found this document useful (0 votes)
3K views

Theory of Equations Exercise 2 Solution PDF

1. The document discusses theory of equations including: - Finding real roots of polynomials - Using Descartes' rule of signs to determine maximum possible real roots - Relationship between discriminant and nature of roots 2. It also discusses properties of polynomials including: - Rationality of values and derivatives at irrational points - Using the rational root theorem to prove irrational roots - Relationship between turning points and nature of real roots for cubic polynomials 3. Key concepts covered are real and complex roots, Descartes' rule of signs, discriminant, rational root theorem, stationary/critical points and turning points.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
3K views

Theory of Equations Exercise 2 Solution PDF

1. The document discusses theory of equations including: - Finding real roots of polynomials - Using Descartes' rule of signs to determine maximum possible real roots - Relationship between discriminant and nature of roots 2. It also discusses properties of polynomials including: - Rationality of values and derivatives at irrational points - Using the rational root theorem to prove irrational roots - Relationship between turning points and nature of real roots for cubic polynomials 3. Key concepts covered are real and complex roots, Descartes' rule of signs, discriminant, rational root theorem, stationary/critical points and turning points.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Theory of equations

1.

p(x) = x 4 + 4x 3 8x 2 1 ,

then p'(x) = 4x3 + 12x2 16x = 4x(x + 4)(x 1)

(a)

Let

x = 4, 0, 1 are roots of the equation p'(x) = 0

p(x)

There is one real root with x < 4 and another with x > 1.
There is no real root with 4 < x < 1 .

Since p(5) < 0 , p(6) > 0

and

There are two real roots

p(1) < 0 and p(2) > 0 .


and 1 < < 2 .

(b)

Let p(x) = 8x 5 5x 4 40x 3 50 , then p'(x) = 40x4 20x3 120x2 = 20x2(2x + 3)(x 2)

x = 3/2, 0, 2 are roots of the equation p'(x) = 0

3/2

p(x)

There is one real root with x > 2.

Since p(3) < 0 , p(4) > 0 ,there is one real root

2.

6 < < 5

for p(x) = 0 with

(a)

for p(x) = 0 with

3< < 4 .

Let p(x) = x 4 + 2x 2 + 3x 1 .
The number of sign change for

p(x) is 1 .

By the Decartes' rule of sign, there is at most one positive root.


p(x) = x 4 + 2x 2 3x 1 .
The number of sign change for

p(x) is 1 .

By the Decartes' rule of sign, there is at most one negative root.


There are at most two real roots.
Since deg[p(x)] = 4, there is at least 2 complex roots.
p(2) < 0 , p(1) < 0 and p(0) < 0 , p(1) > 0 .

There are 2 complex roots and 2


2 < < 1

real roots

for p(x) = 0 with

and 0< < 1 .

)(

(b)

Let p(x) = x 5 2x 3 + x 10 , then p'(x) = 5x 4 6 x 2 + 1 = 5x 2 1 x 2 1

x = 1,

,1

p(x)

are roots of the equation p'(x) = 0

There is one real root with x > 1.

Since p(1) < 0 , p(2) > 0

,there is one real root

for p(x) = 0 with

1< < 2 .

3.

, , are the roots of the equation x3 px + q = 0


++=0

+ + = p

(2)

( )2 = 2 + 2 2 = 2 + 2 + 2 2

Now,

= 0 2 2( p ) 2

( )2 , ( )2 , ( )2

y = 2p x 2 +

.
.

(1)
= q .

(3)

(4)

2
2
2
= ( + + ) 2( + + ) 2

2( q )
2q
= 2p 2 +

, by (1) - (4) .

are roots of a new equation formed by putting:

2q
3q
. (5)
xy = 2px x 3 + 2q x 3 px + q + xy px = 3q xy px = 3q x =
x
yp
3

3q
3q

p
+ q = 0 y 3 6py 2 + 9p 2 y + 27q 2 4p 3 = 0
yp
yp

Substitute (5) in (1),


The product of roots of

D = ( )2 ( )2 ( )2 = (27q2 4p3) =

(6)

4p3 27q2

.(6)
. (7)

Since (1) is of degree 3 , the condition of roots are shown below:


(i)

3 real and distinct roots

(ii)

1 real double root and one other real root or 1 triple root D = 0 .

D>0.

(iii) 2 complex roots and 1 real roots


(w.l.o.g. let

4.

The roots of (1)

D<0.

= a + bi, = a bi , = c and substitute in (7) )


D0

should be real

4p3 27q2 0 .

x 2 (3 + 2 )x + (4 ) = 0

x2 3x + 4 = (1 + 2x)

This quadratic equation has real roots

= (3 + 2 ) 4(4 ) 0
2

42 + 16 0

4 21
2

or

4 + 21
2

x2 3x + 4 = (1 + 2x)

y = x 2 3x + 4 ....(1)

y = (1 + 2 x ) ....(2)

(2)

is a pencil of straight lines passing through

the point

=3

1
,0 .
2

4 21
2

4 + 21
2

(1) and (2) cut each other only when

4 21
2

or

4 + 21
2

1
,0
2

as shown in the graph.

5.

When

= 3,

When

9 77
9 + 77
<x<
2
2

(2) becomes

Let y = (x 1)2 ( x a) + t

y = 3 + 6x .
(0.11 < x < 8.89),

3 + 6x greater than (x2 3x + 4) .

(a > 1)

y' = (x 1) (3x 2a 1) , y" = 2(3x a 2)

y"

y"

x =1

= 2(1 a ) < 0 since a > 1.

y is a max. when x = 1,

2a + 1 = 2(a 1) > 0 since a > 1.


x=
3

As x -, y -

y is a min. when x =

4(a 1)
+t<0
27

ymax > 0 ymin < 0

4(a 1)
27

t>0

ymax = t.

2a + 1
,
3

4(a 1)
+t
27
3

y min =

as x +, y + .

and

The equation y = 0 has three real roots

2a + 1
.
3

x = 1 or x =

For critical values, y' = 0 .

3
t = 4(a 1)

27

0<t<

y = (x 1)2 ( x a) + t y = x3 (2 + a)x2 + (1 2a) x + (a + t)

++=2+a

+ + = 1 2a
= a + t

+ + - 2( + + ) + 3
t=0

= 1 2a 2 (2 + a) + 3 = 0

6.

Let x =

p
q

be a rational root of ax2 + bx + c = 0 , where p, q Z , H.C.F (p,q) = 1, p , q 0 .

By the rational zero theorem, p|c and q|a .


Since a, c are odd and p , q 0 , p, q are odd .
2

However

p
p
a + b + c = 0
q
q

ax2 + bx + c = 0

ap2 + bpq + cq2 = 0 (q 0)

There is a contradiction since ap2 , bpq, cq2 are all odd so their sum is also odd,
and must not be equal to 0 .
7.

(a)

P(x ) a 0 x n + a 1 x n 1 + ... + a n 1 x + a n

P(p) P(q) = a 0 p n q n + a 1 p n 1 q n 1 + ... + a n 1 (p q )

[ (

= (p q ) a 0 p

n 1

+p

n 2

q + ... + q

n 1

) + a (p
1

n 2

+ p n 3 q + ... + q n 2 + ... + a n 1

Since a0 , a1 , , an , p, q are integers, the second factor in the above expression is an integer.
The first factor (p q) is an even integer if

(b)

P(p) P(q)

p, q are both even integers (or both odd integers).

is even .

Suppose, on contrary, x = q is an integral root of P(x) = 0. Then P(q) = 0 .


(i)

If q is odd, then P(1) P(q) is even by (a) .


But P(1) P(q) = P(1) 0 = P(1),

(ii)

which is odd by given.

If q is even, then P(0) P(q) is even by (a) .


But P(0) P(q) = P(0) 0 = P(1),

which is odd by given.

In both cases, there is a contradiction.

8.

If

is the root of the equation

ax2 + bx + c = 0, then a2 + b + c = 0 and a2 = b c .(1)

A2 + B + C is rational a(A2 + B + C) = A(b c) + Ba + Ca , by (1)


= (Ab + Ba) Ac + Ca is rational
Ab + Ba = 0, since is irrational and all other constants are rational .
Ab = Ba
3

a = b c

From (1),

(2)

(3)

A3 + B2 +C is rational Aa3 + Ba2 +Ca = A(b2 c) + Ba2 +Ca , by (2)


= (Ba Ab) 2 + (Ca Ac) = (Ca Ac)

is rational , by (3)

Ca Ac = 0 , since is irrational and all other constants are rational .


Ac = Ca
Let p(x) = a 0 x n + a 1 x n 1 + ... + a n 1 x + a n .
2

p(x) = (ax + bx + c) q(x) + mx + n, where


Now, p(x)

Then by Division Algorithm,


m, n

are rational constants.

is rational (a2 + b + c) q() + m + n

is rational .

m + n is rational , given that a + b + c = 0


2

m = 0, since is irrational
p(x) = (ax2 + bx + c) q(x) + n , that is, the remainder is independent of x .
9.

Let y = p(x) = x3 px + q then p(x) = 3x2 p, p(x) = 6x


Since

lim p( x ) = + , lim p( x ) = ,

x +

p(x) = 0 has at least one real root .

For stationary points, p(x) = 0, 3x2 p = 0


From (1),

x=

p
3

(1)

There are two turning points if p > 0.

Since

p
p" +
>0

y is a min when x = +

p
.
3

y min = q

2p
3

p
,
3

Since

p
p"
>0

y is a max when x =

p
3

y max = q +

2p
3

p
,
3

p(x) = 0 has 3 real roots (or 1 double real root and 1 real root) ymin 0 and ymax 0

4p 3 27q 2

2p p
0
3 3

and

y has only one real root

2p p
2p
q+
q

3 3
3

(a)

q+

2p p
0
3 3

2p p
2p p
q+
0
q

3 3
3 3

[ymin > 0 and ymax > 0] or [ymin < 0 and ymax < 0]

p
>0
3

4p3 < 27q2 .

x3 2x + 7 = 0, Since 4p3 = 4(2)3 = 32 , 27q2 = 27(7)2 = 1323, 4p3 < 27q2.


The equation has only one real root.

(b)

3x3 + 4x 2 = 0, Since 4p3 = 4(-4/3)3 = -256/27 , 27q2 = 27(-2/3)2 = 12, 4p3 < 27q2.
The equation has only one real root.

(c)

4x3 7x + 3 = 0 . Since 4p3 = 4(7/4)3 = 343/16 , 27q2 = 27(3/4)2 = 243/16, 4p3 > 27q2.
The equation has 3 real roots .

10.

(a)

x3 + Px2 + Qx + R = 0

(1)

(2)

Put x = X + k , where k is a constant .


(X + k) 3 + P(X + k)2 + Q(X + k) + R = 0
2

In (2) , the coeff. of x -term is (3k + P).


By putting 3k + P = 0 or k =

P
P
, then the transformation x = X
3
3

can transform (1)

into (2) where x2-term disappears.


(b)

x3 15x = 126

(3)

x=y+z

(4) (3), (y + z) 3 15(y + z) = 126

y3 + z3 + 3y2z + 3yz3 15(y + z) = 126

y3 + z3 + (3yz 15) (y + z) = 126

(4)

y3 + z3 + (3yz 15)x = 126 .

(5)

If further we choose 3yz 15 = 0 , or yz = 5

(6)

(7)

(8)

(9)

y + z = 126

(5) becomes :

From (6), (y3 ) (z3 ) = 125


3

From (7) and (8), y

and z

are roots of a quadratic equation t 126 t + 125 = 0

If y3 and z3 are both real and y0 , z0 are their real roots, possible values for y , z are y0,
y0, 2y0

and z0, z0, 2z0 .

By (6) , the product yz must be real. Therefore the only combinations consistent with this give, as
y0 + 2z0 ,

the three roots of the cubic, by (4) : y0 + z0 ,

2y0 + z0 .

Similarly for the eq. x3 15x = 126 .


From (9) ,

Then the solutions are


11.

(a)

Take y03 = 1,

t = 1 or 125.

x2 + px + q = 0

1 + 5, + 52 ,

z03 = 125, we have

y0 = 1,

z0 = 5 .

2 + 5 .

+ = p , = q .

(1)

(3 )(3 ) = 33 + (4 + 4) = ()3 + [( + )4 4(2 + 2) 622]


= ()3 + {( + )4 4[( + )2 2] 6()2} = q3 + q {(p)4 4q[(p)2 2q] 6q2} , by (1)
= q3 + 2q2 + q (p4 4p2 q + 4q2) = q(q + 1)2 (p2 2q)2

(2)

One of the roots of the equation x + px + q = 0 is equal to the cube of the other

(b)

3 =

(p2 2q)2 = q(q + 1)2 .

or

x2 + px + q = 0

3 =

(3 )(3 ) = 0

+ = p , = q

q(q + 1)2 (p2 2q)2 = 0 , by (2)

(1)

(t + ) + (t + ) = (1 + t) ( + ) = (1 + t) (p) = p(1 + t)
(t + ) (t + ) = t(2 + 2) + + t2 = t[( + )2 2] + + t2 = t[(p)2 2q] + q + t2 q
= t2 q + t(p2 2q) + q
Hence the new equation is

x2 + p(1 + t) x + t2 q + t(p2 2q) + q = 0

If p2 4q < 0 , p 0 , then (1) has two complex conjugate roots and .


Since p, q are real, and and are complex conjugates ,

(1)

If t 1 and is real, (t + ) , (t + )

(2)

If t = 1 , (t + ) , (t + )

are complex numbers.

are equal to p .

Since p 0 , therefore (t + ) , (t + ) are different from zero.


12.

(a)

p2 and q are the roots of x2 + bx + q = 0

p +p b+q=0

(2)

Case (1), If q 0 , then from (3) ,

. (1)
2

q + bq + q = 0

(3)

q + b + 1 = 0 q = (b + 1)

(4) (2), p + p b + (b + 1) = 0
4

p2 = (b + 1) or p2 = 1

p = (b + 1) or

(4)

(p + b + 1) (p 1) = 0
2

p = 1 .

Case (2), If q = 0, then (2) becomes p4 + p2b = 0 p2 (p2 + b) = 0

(b)

p = 0 or

The equation x2 + bx + c = 0

+ = b , = c

(1)

( + ) 2 + 2 = b 2c + 2
1
1
+ +2
+
=
=
2
2
1 + 2 1 + 2
1 + 2 1 + 2
1 + b 2 2c + c 2
1 + ( + ) 2 + ( )
2

)(

1
1
1
=
1 + 2 1 + 2 1 + b 2 2c + c 2

If

The new equation is

(1 + b

2c + c 2 x 2 + 2c 2 b 2 x + 1 = 0 .

b = 1 and c = 1 , then the given equation is x2 + x 1 = 0 , with roots

x=

The new equation is 5x2 5x + 1 = 0, with roots

13.

, , r
r

Let

Then + 1 + r = 3
r

b
3

(3)/(2),

(6)(1),

x 3 + 3x 2 + bx +

x=

b
or
3

x3 + 3x2 + bx + c = 0

be the roots of

. (1)

(2)

2 +1+ r = b
r

(5)

(5)(4),

c=

From (5)

and division,

3 3
3

x=
2

1 5
2

5 5
.
10

b2
= 0.
27

x=

b2
27

(3)

(6)

3 = c

(4)

b 2
b
b2

x + x + 3 x +
=0
3
3
9

b
4b 2

3
9

(7)

(8)

In (7),

b 4b 2
b 2 4b 2 27 6b b 2

= 9 2b +

=
= 3
3
9
9
9
3

From (5), Since


In (8),

is an integer and 0 , b

must be a non-zero integer divisible by 3 . (9)

Since the roots are integers, > 0

b2 + 6b 27 = (b + 9)(b 3) < 0 9 < b < 3

(10)

The only b satisfying (9) and (10) is that b = 6 and = 9 .


From (7),

14.

x = 2, 1, 4.

x4 + a3 x3 + a2x2 + a1x + a0 = 0
4

z is a root of (1),
1/z is a root of

(1)

(2)

= 0

z + a3 z + a2z + a1z + a0 = 0
(1/z)4 + a3 (1/z)3 + a2(1/z)2 + a1(1/z) + a0 = 0

(1),

a0z4 + a1z3 + a2z2 + a3z + 1 = 0

a
a
a 1
z 4 + 1 z 3 + 2 z 2 + 3 z +
a
a
0
0
a0 a0

(3)

Since z and 1/z are roots, (2) and (3) are identical.
a0 =

By comparing coefficients of constant terms of (2) and (3),

1
2
a 0 = 1 a 0 = 1
a0

(by comparing coeffs of x2-terms)

Case (1),

When a0 = 1, a1 = a3

Case (2),

When a0 = 1, a1 = a3 , a2 = 0 (by comparing coeffs of x3-terms

The necessary and sufficient conditions are : [(a0 = 1,


4

a1 = a3)

and x2-terms)

or (a0 = 1, a1 = a3 , a2 = 0)]

Put p(x) = x + a3 x + a2x + a1x + a0


For Case (2),

p(1) = 1 + a3 + a2 + a1 + a0 = 0 ,

p(1) = 1 a3 + a2 a1 + a0 = 0 .

(x 1)(x + 1) is a factor of p(x) by Factor Theorem.

Division of p(x) by (x 1)(x + 1) gives a quadratic factor and the problem reduces to solving quadratic
equation.
For Case (1),

equation

x 2 + a1x + a 2 +
Put

y=x+

(1)

reduces to x4 + a1 x3 + a2x2 + a1x + 1 = 0

a1 1
1
1

+
= 0 x 2 + 2 + a1 x + + a 2 = 0
x x2
x
x

1
x

(4)

(5)

(6)

(7)

(1)

1
1
x + 2 = x + 2 = y2 2
x
x

Then

y2 2 + a1y + a2 = 0 or y2 2 + a1y + ( a2 2) = 0

(4)

becomes

(6)

is a quadratic and let y1 and y2 be the roots. From (5), we have:

x y1x + 1 = 0 and x2 y2x + 1 = 0


which are quadratic equations in x

15.

and (7) gives the roots of (1) .

x4 s1x3 + s2x2 s3x + s4 = 0


Writing as symmetric sum, then
= s1

. (2)

= s2

(3)

, = s3

( + ) = = s2
( + )( +) = 2 = ()() 4 = s1 s3 4 s4

(4)

(6)

= s4

(5)

(7)

( + )( +)( +) = + = () + [() 222]


3

2 2 2

= ()[()2 2 ] +[()2 2 ] = s4 [s12 2s2] + [s32 2s4 s2] = (4s2 s4 s12 s4 s32).(8)
From (6), (7), (8), the required cubic equation is

y3 s2y2 + (s1 s3 4 s4)y + (4s2 s4 s12 s4 s32) = 0

(9)

Supposing that methods of solving cubic equation are known,


(9) is solved and let the roots be y1, y2 and y3 . Then:
+ = y1
From (5),

+ = y2

(10) ,

()() = s4

(11)

+ = y3

are roots of the quadratic equation : z2 y1z + s4 = 0


z1 = , z2 =

Suppose (14) is solved, let the roots be

From (3), ( + ) + ( + ) = s3 z2 ( + ) + z1 ( + ) = s3
( + ) + ( + ) = s1

From (2),

(12)

(13)

and

From (10) and (13),

(15)

(16)

(17)

(18)

(14)

(17) and (18) are simultaneous equation with unknowns ( + ) , ( + ) .


k1 = + , k2 = +

Let the equations be solved and

(15) , (16) form two quadratic equations for which , , , can be found.
Note that if (11) or (12) is used instead of (10) , the solution is the same due to the arbitrary use of the
, , , for the four roots of (1).

notation
16.

(a)

We use induction on k that xk + x-k can be expressed as a polynomial

pk(z) in z = x + x-1 .

The proposition is obviously true for k = 1 where p1(z) = z.


For

x2 + x-2 = (x + x-1)2 1 = z2 2

k = 2,

Assume the proposition is true for k n


For

(n+1)

k = n + 1, x

-(n+1)

+x

where n

-n

-1

.
(n-1)

= (x + x )( x + x ) [x

= pn(z) p1(z) pn-1(z)


Putting

and the proposition is true for k = 2.


+ x-(n-1)]

, by inductive hypothesis.

pn+1(z) = pn(z) p1(z) pn-1(z) , then the proposition is also true for k = n + 1.

By the Principle of Mathematical Induction, the proposition is true k

(b)

x 6 + ax 5 + bx 4 + cx 3 + bx 2 + ax + 1 = 0
3

-3

-2

-1

(x + x ) +a(x + x ) + b( x + x ) + c = 0
Putting z = x + x-1 , then (1) becomes

(c)

z3 = az2 + (b 3)z + (c 2a) = 0

If is a root of the polynomial equation (1), then

Putting z = x + x , then (4) becomes


or

z2 + 4z + 3 = 0
-1

....

(2)

(3)

is a root of (2) .

(4)

(z 2) + 4z + 5 = 0

(z + 3)(z +1) = 0
or

z = x + x-1 = -1

x2 + 3x + 1 = 0

or

x2 + x + 1 = 0

or

x=

3 5
2

(1)

z = x + x = -3

x=

+ 1

(x2 + x-2) + 4( x + x-1) + 5 = 0

-1

....
(z3 3z) + a+ bz + c = 0 .

or

x 4 + 4 x 3 + 5x 2 + 4 x + 1 = 0

1 3i
2

17.

(r 0) .

x3 + 3qx + r = 0
++=0

(1)
+ + = 3q

(2)

Put

p(x ) = r 2 x 2 + x + 1 + 27q 3 x 2 (x + 1)

Then

p = r 2 2 + + 1 + 27q 3 2

= r

(3)

(4)

2
2
2 + +1

3 ( + )

+ 1 = r 2
+
27
q

2
2
2

2 2
3 3
( 3q + r ) + ( 3q + r )
2
3
+ 27q 3 4 ( )2 = r 2

+
27
q
= r 2

2 ( )
4

( )
3

3q
( )2 = 27q 3 r 2 + 27q 3 ( r )2 = 0
= r 2 2 + 27q 3
6
6
6

3

satisfies

(a)

When q = 0 ,

p(x ) = r 2 x 2 + x + 1 + 27 q 3 x 2 (x + 1) = 0
3

x3 + r = 0

(1) becomes
3

The roots of (6) are

r,

r ,

Take be any one of these three roots and


=

Then

When q = 0 ,

= or

, where

(6)

be any other root not equal to .

(5) becomes

(5)

is the complex cube roots of unity.

or

x2 + x + 1 = 0

(7)

(r 0) and the roots are , 2 .

= 2

(8)

(9)

(10)

It can be seen that (7) and (8) are the same.


(b)

If

4q 3 + r 2 = 0 , then

(9) (5),
If

r2 = -4q3

p(x ) = 4q 3 x 2 + x + 1 + 27 q 3 x 2 (x + 1) = 0
3

q = 0, then by (9) , r = 0 , contradicting to r 0 .

q 0 and (10) becomes

4x6 +12x5 3x4 26x3 3x2 + 12x + 4 = 0

(i)

If

=1, then = .

(ii)

If

= -2, then = -2 . From (1),

(iii) If

= , then
2

= 1 or -2 or

satisfies (10),

Since

4 x 2 + x + 1 + 27 x 2 (x + 1) = 0

(x 1)2 (x + 2)2 (2x + 1)2 = 0

1
.
2

-2 + + = 0 ,

= -2 . From (1), -2 + = 0 ,

=.
=.

In any one of the above cases, there is a double root for (1).

18.

(a)

A.M. G.M. : Let xi (i = 1, 2, ., n) be n distinct positive numbers, then

(b)

f (x ) x n + a 1 x n 1 + a 2 x n 2 + ... + a n 1 x + a n = 0

1 = a 1

Then

has n distinct positive roots i (i = 1, 2, ., n)

1 2 = a 2

(1) ,

n 1
1 2 ... n 1 = ( 1) a n 1 . (n-1)
in i
a i = ( 1) b i , then
i

Writing

(2)

, .,

1 2 ... n 1 n = ( 1) a n . (n)
n

nn
n
n
n
a n = ( 1) b n = ( 1) b n
n

n 1 n
b n 1 n 1 = ( 1)n 1 nb n 1 n 1
a n 1 = ( 1)
n 1

From (i) and (n),

b n = 1 2 ... n 1 n

From (ii) and (n-1),

nb

n 1
n 1

= 1 2 ...

n 1

b n = (1 2 ... n 1 n )

1/ n

From (iii),

xi
> x i .
n

(i)

(ii)

(iii)

(iv)

1`
1` n 1

= (1 2 ... n 1 ) n 1 = (1 2 ... n 1 ) n

1 2 ... n 1 n 1
n 1
<
< b n 1
n

1
n 1

= b n 1

(v)

n 1 n 2
2
1

f ' (x ) nx n 1 + (n 1) x n 2 + ... + 2a n 2 x + a n 1 = n x n 1 + (
) x + ... + a n 2 x + a n 1
n
n
n

Consider f '(x) = 0.
Since

f(x) = 0

has n

distinct positive real roots and between any two roots of

f(x) = 0 , there is at

least one real root of f'(x) = 0, where deg f'(x) = n 1 .

f '(x) = 0

has

(n 1) distinct positive real roots, each lying between two roots of

Let 1, 2 , ., n-1 be the (n 1) distinct positive real roots of


n 2
1 2 ... n 2 = ( 1)

1 2 ... n 1 = ( 1)

n 1

2 n (n 1)
2
n2
n 2 2
n 2 n
n 2
b n 2 n 2 =
= (n 1)b n 2
a n 2 = ( 1)
b n 2
( 1)
n
2

n
n
n
2

b n 1 = (1 2 ... n 1 )

f '(x) = 0

a n 1
n 1 1
( 1)n 1 nb n 1n 1 = b n 1 n 1
= ( 1)
n
n

1
n 1

1
n 2

f(x) = 0 .

(vi)

(vii)

(viii)

1
1

n 1
n 2
= (1 2 ... n 2 ) n 2 = (1 2 ... n 2 ) n 1

(n 1)b n 2 n 2 n 2
1 2 ... n 2
= b n 2
=
<

n 1
n 1

Similarly, by considering f"(x), f'''(x), , f(n-2)(x), we get correspondingly,


bn-2 > bn-3, bn-3 > bn-4 , , b2 > b1 ,
Combining all these inequalities, we have b1 > b 2 > ... > b n 1 > b n .

10

You might also like