are for Torm of theorems

known as Mean
In this chapter we shall
properties
theorems. deal with
theorem, Rolle's
mean value theorem,
Cauchy's mean value
and their ge theorem,
'stheorem
LAgrange's
ometrical
Taylor interpretations and
pplications.
Rolle's
Theorem.
92. [W.B.U. T 2011]
a function
Letf beea
defined on a finite closed
interval [a, b] such
() f) is continuous for all x in
asxSb.
that

(Gii) f'() exists for all x in a <x<b

and (iii) (a)=f(b).
Thenthereexist at least one
value c, a<c<b, such that
f'(c) =0.
Proof. Beyond the scopeof the book.
Ceometrical Interpretation of Rolle's
Theorem.
A function fwhich is continuous on
the closed interval [a, b], can
be represented graphically by a curve without break (see the
following
fo), Since fis derivable on (a, b), the curve has a tangent at each
on the arcAB where
point A= (a, f(a) and B= (b, f(b).
Also,since f(a) = f(b),so the ordinates of the two points A and
B are same.

A B
y= f(r)
f(a) f(b)

X=a X=C
Now f(c) =0means the tangent at the point C to the curve is

parallel to the x-axis where C =(c,f(c).

202 ENGINEERING MATHEMATICs-I

Hence the geometrical of Rolle's theorem is "Ifthe

interpretation
graph ofy=f(r) is represented by the arc AB without any
break on
[a b]having a tangent at every point and ordinate of A
and B points
are same,then there exist at least one point
C, on the arc AB, where
the tangent is parallel to the x-axis."
9.3. Illustrative
Examples.
Example. 1. Verify Rolle's theorem in each of the following cases :
() f)=cosxin
(i) fo) =2r'+x-4x-2, -/2sxs2
(ui) f)=2 +(r-Iý3, 0sxs2
(iv) fo)x, -1sxsl. [W.B.U.T. 2012,2003]
Solution:(i) We know the function
cos x is continuous and
for all real values ofx. derivable

Now
f()is continuous

f'(x)=-sinx,
-<x<.
2
on-g and derivable on
(
Also

Thus f(x)satisfies all the conditions of Rolle's

theorem.
By Rolle's theorem,we should have

fc)=0where
-<o<
2
 whose two solutions are
(3c-3 C=-1,
3 and
2(c+l)
eas wellas
-v2 3 <<5
is verified for the
theorem given function.
Rolle's 2
Hence,

dere f(x) =2+(X-)° iscontinuous on 0sx<2 hut

does not exist at x=1.

f(x) is not derivable

on 0 <x<2.
Hence

of Rolle's theorem do
the conditions not hold. So Rolle's

is not applicable to the given function.

theorem
given function can be written as
(i) The
f(x)=-x, when -1<xs0
=X, when 0 <xs1.

Obviously f(r)is continuouson [-1,1].

+ -h-0 =-1
Now Lf'(0) =h0
lim
f(O h)-f(O) =h0-
lim
h h

f0+h)–f(0) h-0
Rf (0)=h0+
lim = lim =1.
h h0+ h

Lf(0) = Rf'(0).
So f'(x) does not exist at x=0 which lies between -1and 1.

Hence fa) is not derivable on -1<x<l.

Thus the conditions of Rolle's theorem do not hold.
So Rolle's
theorem is not applicable to the given function.
 Again f(0) = 0= f(2)

satisfies all the conditions of Rolle's theorem. Then

st one point x =con 0<x<2, such that f(c)=0

S6-24
6 3
1.732
3 1577,1423
Both these values ofc lie in (0,2).

Example 3. ()Is Rolle's theorem applicable to the function

f()=1-x in (-1, 1]? Justify your answer.

1

G)IsRolle's theorem applicable to f()=in

2-x? (-1, 1]?
Justify your answer.
2

Solution :(0) Here, f(x) =1-x.

If x* 0,f(r) =-2 3
1

3x3
So, f(0) does not exist, and, -1<0<1, ie. f(r)is not derivable
in (1, 1)
2
Hence Rolle's theorem is not applicable tothefunction f(*) =l-x3

)
on the interval

S)=-2
S(1) 2-x2
-1,1].

r)=(-0(2-)x(-2x) =
1

is continuous in (-1, 1.
1):

2x
which existsin (-1,1).
(2
1 1

Also, =2-1 =|and f(-1)==1

f(1) 2-1
So, f()=f(-1)
 LAWS
ENGINEERING
MATHEMATICS-]
Solut
206 of Rolle's theorem.
allthe
conditions Hence
1

Here, fx)
satisfies onthe
J(*)=,-y2 interval-1,1.
is applicable
to
Rolle's theorem
Rolle's
theorem is applicable tothe theot
whether
4. Explain containing the origin.
Example interval
in any closed
x)=x
function
|-a, a Where a > 0, containi
Exa
Solution :Let us consider an interval

the origin. as
The function r)=x can be re-written
inte

r)=x, when x>0 Sol

=-X, When x<0

x=0, since
Here, x) is continuousat

lim o(r) =X0

I0+ lim x)=0)=0.
h-0 =1
o(0+h)-(0)= lim
But Ro'(0) = lim h0t h

(0+h)-o(0) = lim -h-0-=-1

and Lo'(0) =h0
lim
h h0 h

Since Ro'(0) =Lo0), x) is not derivableat x=0. So, x) is not

theopen interval (-4a).One of the conditions of Rolle's

derivable in

Theorem is not satisfied.Hence Rolle's theorem is not applicableto the

function o(:) = in any interval [-a,a] containing the origin.

Example 5. Does f(x) =cos satisfy Rolle's theorem inthe interval

-1sxs1? Justify your answer.

Solution : Since f(*) =coss0) is undefined.

So, fx) is not continuous at x=0.
Thus f(x)does notsatisfy, the first condition of Rolle's theorem
in (-1, 1].

Hence theorem the

Rolle's is not applicableto f()= cos
in

interval [-,1].
Example. 6. Show that Rolle 's theorem is not applicable to
fo)=tanx in [0,z], although f(0) =f(r).
 OF MEAN
LAWS 207

solution
:Here f) = tan x is
continuous every where in [0, n]
except
and consequently it is not derivable
atx= there.

the condition Rolle's

theorem do not hold.

of
Thus Hence Rolle's
is to the function
not applicable f(x) =tan x on the
theorem interval
although f(O)= f(r).
[0.),
7. Test the applicability of Rolle's theorem
Example for the function
dwl=(r-a)"(-b)" in asxsb, where m and nare positive

integers [MAKAUT 2018]

Solution : Here, f(*)=(- a)" (*-b)'

Since m and n are positive integers, f(x) is a polynomial ofdegree
Im+n),which is continuous atevery point. So f(x) is continuous in

asxsb.
Now, f'()= m(x-a"(*-b)" +n(x-a)"(*-b)'l
-(*-a)(-b){me-b)+ n(r-a)} ... (1)

which obviously exists in a<x<b

Also, f(a) =0=f(b)

Thus fx)satisfies all the conditions of Rolle's theorem Then there
cin a<x<b, such that f(c) =
0
isat least one point x =

From (1), then we have

+nfo-a)} =0
(c-a)(e-b)"fmo-b)
Since a<csb
c-a#0, c-b# 0

.mc-b) +ne-a)=0
mb +na
c= m+n
(m+n)c=mb
i.e,
+ na
or,
in theratio m:n.
theinterval a, b] internally

The pointc divides to show

that,

function f(r)=(*-2)logx, lying

Consider the value of xr

Example 8. by at least one

logx=2-x
is satisfied
x
heequation
betweenl and 2.

Solution : Here, f()=(r-2)logx

 -2,
logr+
ra)=
in (1,2) and so continuous in [1.
exists derivable 2],
Sa) f(x)is
function of Rolle's
theorem,there exists at
Thusthe conditions
c,such that
all
As (x)satisfies (l,2), say
of x, lying in
value
least one
I<c<2
rc)=0,
we have clogc
+c-2=0
From (1)
or, clogc=2-c
equaions
which shows that the
x logx =2-x
by c where <c<2.
1
is satisfied
is proved.
Hence the proposition
that between
any two roots of e cosx = 1, there

Example 9. Show
of e'sin x-|=0
at least one root
exists
of the equation
distinct roots
Solution :Let aand ßbe any two
e* cosx =1
(1)
Then e cosa =1= e cosB

Let us construct in a function f defined by

(2)
-COSX

in the closed interval asxsB.

in a,B,since cosx and eare
We note that f(x)is continous
continuous in that interval.
(3)
Also fx)=-e* + sinx, where a<x<ß
So, f()is derivable is (a,B)

= 1-e
COSC
From (2), f(a) =e -C0S =0 [Using (1)]
e
Similarly, f(B) =0
So, f(a) =f()
Thus the functionfsatisfies all the conditions of Rolle's theorem
in (a,B]
 209
LAWS
OF MEAN

exists at least one value of x,

So, there

say x=Y,
a<y <B
such that
fY)=0
siny =e 0
or, from (3),

i.e., e siny -1=0

B)is a root oftheequation e sinx - =0
1

This proves that y E(a,

any two roots oa, B (a #B)
of the
proved that between
Thus it is the equation
=1, there exists at least one root of
eJuation e cosx

e sin x-l=0.
Value Theorem.
94.Lagrange's Mean [a, b] such that
on a finiteclosed interval
fbea
Let function defined
asxsb
0 fx)is continuous for

x,
all x,

a<I<b.
() f(x)exists for all

c, a<c<b, such that

exist at least one value
Then there
[WB.U.T. 2004)
fb)- f(a) -f(o).
b-a
o() defined by
a new function
Proof. Let uus construct (1)
r)=f(x)+ Ax,
=

)
o(a) d(b)
in such a
way that
chosen
where2. is
a constant
f(b)+
= Ab
from (1) f(a)
+ha

,
Then

or A(a-b) f(b) =
-f(a)
(2)

i.e.,
2. = a-b
b-4
214

Example. 3.Ifaand b are distinct real

ENGINEERING
MATHEM
MATICS-1
S OFMEAN
5.Iff
numbers, show
areal number c between a and b such that that
there S(r)
that
a' + ab +b? =3c2
exists
:Let
Solution : Let us considerthe function stion

f(o)=r' defined on [a, b].

applyi
Then
S)=3r
Obviously fx) is continuous Heget,
son [a, b] and f'(*)
in
existsin
So, Lagrange's Mean value theorem (a,b).-
interval [a,b]
can be applied
to
f)
forthe Hence
f(
Hence there exists a
number c, a<c<b, x
Since

such that
b-a =f(c), a<csb t f(r)
is

6
Lanple.
or
b-a -=3e? (b*a) sing mean

Hence a +ab +b? =3c. b).

Example. 4. Prove that olution

for any quadratic
function px + qx +r,the I<X <
value of 9 in Lagrange's
Mean value theorem 1
is always
whatever,
P.9,r,a,h may be. 2 We ge

Solution: Let
f()=pr? +qx+rin [a+h,a
f'()=2px +q
being a polynomial is continuous on [a+h,h) and fx)exists
+
in (a h,h)
But

So,there exists
a real number
(0<0<1)
such that
f(a+h)-f(a) - hf (0h)
He

or,pla+hj +qla +h)+r-pq' -qa-r=h2pla+eh)+9)

or, 2pah ph² + +
gh =2 pah +gh+2ph²0
or,ph² =2phe, ie., 9=-(ph=0)
 OF MEAN 217

9, Use Mean-value theorem to prove the

following
ample.

ixqualities

[MAKAUT 2018,20201]

<log (l + x) <xifxs 0.
[MAKAUT 2018,2020]
1+x

:(0)Let
f(r) =et,
solution

Then from the mean value theorem

f(r)= f0)+x f(0x), 0<0<l,

we have e =e+xe [:fx) = e]

or, e*-1=xetr
e-1
or,
er e-] or, x= log

e-1
or, =-log

0<log e-ll [::0<0<l}

() Let f(r) =log (1 +x).

mean value theorem,

Then from the

fa)=fO)+ xf'
(), 0<@<l,

+x&r)
1

we have
+)=logl
log(l (1)

log (1+x)= 1+
x> 0, so 0<0x<x.
Since 0 <e<land

or, 1<1+&<l+X.

oT, 1> 1+x

1+A

Or, x>+A 1+x

218
ENGINEERING

or,
*<x.
1+x1+A
Tlog(l+x)
1+x <x, by (1).
9.6. Cauchy's Mean Value Theorem.
Let fandg betwo functions
defined on [a, E
) fand g are continuous
on [a, b]
() fand g are derivable
on (a,b), and
(ii) g' ) 0 forany x in
(a, b).
Then there exist
at least one
value c, a<c<

,
f(b)-
f(a)_f)
g(b)-gla) g'c)
Proof.We
construct a new function
define

where 2.isa
c
So J(*)=x', then f(x)= 3x
and g(x) =x +3x,then g'(x)= 2x+3.
We observe that
() J()and g(x) are both polynomial functions and so are
continuous in -1<xsl

(ii) f) and g'() existin -1<xsl

(iii) g'() = 2x+ 3a0 in -1<x<l
Thus the functions f(x) and g(x) satisfy all the conditions of

chy's Mean value theorem. Then there is at least one point x=c in
uch that

3e2
or,
1-(-)
4-(-2) 2c+3

Or.
6 2c+3

2+V112 1+2\7
or, 9c - 2c -3=0, giving C= 18

Both these vaues of care in (-1,1)

Example. 3. In Cauchy's Mean Value theorem,

if f(r)=e and g(x) =e,
1

O is independent bothx and h and is equal to [WBUT 2003]

show that of
222 ENGINEERING MATHEMATICS.1

Solution. Cauchy's Mean value theorem is

fx+h)-f()_ fx+ h)
9 0<0<1.
g(x+h) -g) gx+h)
erth -e*
[-:f()=e,g'(r)=-e*]
e x+h) -e-* -e-(x+9h)

e*(eh-1) e.eh
or,
-0h
e(e -1) -e .e

or,
e"(e" -1)_e or, -eh =-e26h
1-eh e-h
1
or, 26h =h 0=
So is independent of both x and h and is equal to

9.8. Generalised Mean-Value theorem :Taylor's Theorem.

[WB.U.T2001
Theorem-1. (Taylor's theorem with Lagrange's formof remainder

)Let f be a function defined on [a, b] such that

the (n-1)th derivative

(i) the nth derivative

fis
fln exists on (a,b).
continuouson [a, b], an

Then there exist at least one value c, a <c<b, such that

1

s(0) =fa)+(b-a)f(a) +(6-a)'f

2!
 ()=f2) +(*-2)f(2)+ 2!
(*-2)mp+0(x -2)},0<0<1.
3!
1

Since JN
-2)=9 1

1
1

+-2)+-25)
log3
2
(-2)
3! {1+2+0(*-2)}"
1
1

_ (r-2)' (r-2)
l+x)=log3+(-2)
3 18 3 (3+9(x-2)}
where 0<0<1.
a finite series (in power of x) in Lagrange's
Example. 2.Expand cos x in
form of remainder.

Solution :Let f(x) =cosx ::f0)=1.

Now, f ()= cos n+x

nn
f =
c0) cos -2
f'(0) = 0,f"(0)=-1,f"(0) =0, f"(0)=1 and so on.
15
 C
ENGINEERING
226
LAWS
MATHDMATICS.
for the function
Maclaurin
's series f()in Nov
Now the is finite
emainder, fornin
form of
Lagrange's
+
S)=f0)+xf(0)
x

-
n! f"(@x),0<6
<\

cosx =1-2! 4! n!
-cos

Show that
Example. 3.

'(r+)² = x?
23
+r'h+; 2.2
3.1 h

2! Jx+
1
=,0 <0<1. Find
h
when x =0.

Solution : Let f(r)

=x?.
3
f)-,s")-1
2'2*
Now the 2nd order Mean Value theorem

fr+h)=
3
f)+hf()
3
++ 2!
states

),0<0<l.
3 3
+h;x?+ 1

(+h)² =x² h)

3
+3 h'
3.1 1
x2
2.2 2!Vx+Oh

Putting x=0, weget h2

3

= 32 9
8 Jeh 64

Example. 4. Using Mean Value theorem, show that

sinx > x-,fo<x<g

6
Solution :Let f(x) = sin x :. f(O) =0
f(x)= cos x
::f(0) = 1

f"() =-sin x
.. f"(0) =0
f"()=-coS x
i. f"(ar )=- cos r.
 227
OF MEAN
WS
S-1 series for the function f(r) of order 3 is
Maclaurin's
the
Now
m in
()=f0)+f(0)+f0)
2!
+f"
+(a),
3! 0<0<.

sinx =0+x.1+.0+(-cos )
2! 3!
(1)
of,
sin x=x
=X--
6
-.COs

<0<1, so 0< <

Since 0 <x<and
2
0 2

or,
0<cos Ox<1.
0.
-cos x >- 6
6
0r, x-cost>x 6
6
sin x> X66 by (1).
+3r+ 5in a se
the polynomial
f)=x-2r?
Lample5. Expand
powers of (x-2).
positive integral
es of
2x?
(x)=x'-
+3x +5
olution: Here,
-4x+3
then, f(r)=3x?
f")=6x-4
f"a)=6 n24.
and f"la)
=0, when
=8, f"()=
6
=7.f")
f(2)= 1l, f(2)
So, can be
= 0,when n24. form
's
of remainder
dnd f"(2) am with Lagrange
Solution
: Here,
f(x)

24(1.
then
)'.r)=21+*)./")-
ra-(1+ "()m 0,when
>s
and 24, fM(o)
r)24,
Sa r(0)=l.
f(0)=4,/"(0)(0)=
possesses
derivatives of every orderforall

apolynomial
So)being Lagrange sform
of
remainder
with after
ofx.
values & theorem
By Maclaurin
we, have, +
fourterms
„/0)+ 3! 21
/"0)0:),
r9)-f0)+x/0)*/0

=1+rx4 2!
+x24
+x12 6
+x + 24
0<8<1,

or,S(a)
+ 6x?+4x'+x.
ie, (1+x) =1+4x

that if -1<x<l,
Example. 7. Show
1

1 5 10
1
-x+Rs
(1+x) =1+-x--x'
3 9 81 243

where R=
880
243 Ja+ a)
14
3
.
 229

LAWS OF MEAN

Solution :Let
fo) =(1+x)',
-1<x<l.
: fO)=1.
1

8
: f(0)=-Z9
3

10
::f"(0) =27

11

14
: f0)=
80
81

14
880 3
(1+*)
243

Now Taylor'sExpansion of f(r) about x =0, with Cauchy's form

of remainder, after 5terms, is

f)=fO)+x(0)+:2!
4

4! )+-01 (),0 <0<1.

10x80 880x
1 14

1 2x x

:.(1+x)³ =l+ix 5.
81 4!243 47(-6)l+a)
3 92! 27 3!

: 1
(0+) =l+3 9 81 243
14
10
-x*+Rs

-9+a)3
880

where R, 243

Example 8. Expand et as an infinite series in powers of x by the use

of Maclaurin's theorem.

Solution :Let f(x)=e, then f"(1)= e.

of every order for every real value
So, fx) possesses derivatives
of x.
 ENGINEERING MMEAN
OF
230 MATHEMATICS. HS
integral values of n
forall positive and
Again
f"(0)=1, ByMaclauri
=0x) =
R,
n!
for 0<0<l.

for 0 <0<L
theorem,
Soby Maclaurin's

tt e0r
e=l+r4 2! (n-)!
n!
() Sin
X= Si

infinite seeries correspondingtothe finite series

(1) is
Now, the
+.....
I+x+t...+ (2
2! (n-1)! n!
ratio test,
By D'Alenbert's

Sin
x=
Ot,
n+1|

lim = lim =0)<1,for all values of x.

Example.

convergentfor all values ofx.

So, the series (2)is or le
igh

converges to ex if. lim R,=0 Solution

Now,the infinite series n0
Ist P
lim R, = lim
I0 n! Let

= lim x

lim ex =0, fa
nn!

since lim =0 and er (0<0<1) is bounded.

Put

- n!

Hence, e' =1+x+ ofx.

+......+ for allvalues
2! 3! n!
Or,

Example9,Expand sinx ofx withremainder

series inpowers
in a finite:

inLagrange's form.
[WBUT2022]
Solution :Let, f(x) =sinx, then f"(x) = sin
+ and
2
f"(0) = sin
nTU

2
So, f(x) ofx. val
possessesderivatives value
of all orders for every
 3T
-sin
+
3!

(n-1)r
-sin
2 n!-sin+0x,
2 o<0<l:
(n-1)!

or, sin x = x- sin (n-) sin +0n

(n-1)! 2 n! 2
-sin

0<0<1..

Example. 10. Prove that sin

46°2+
46°
2 30
Is this estimate

high or less.
[WB.U.T 2003, 2022]
Solution :We show two different
processes.
IstProcedure (Using
M.V. T) :
Let f(x)=sin x. By Lagrange's M.V.T. (2nd form) we have
f(a+h) =f(a) + hf'(a+ eh) where 0 << 1

fla+h)-fla) =hf(a+0h).
Putting a= 45°,h=|° we get from above
f(46°)=f(45°)+|°x cos (45°+ 0x 1° )
or, sin 46° = sin 45° +cos(45°+
180
6°) ) (1)

sin 45°+ -cos45°. [: is so small]

180

sin46° = I+
180
.. sin 46°
180

For small , cos +

(45° )<cos45°. Therefore from (1), exact
of sin 46° < Approx value of sin 46°.
of Value

This shows that the estimate is high.

232 ENGINEERING
MATHEMATICS -I
2nd procedure (using Taylor 's Theorem) :
Let fx) =sin x.Applying 2nd order
Taylor's Theorem we get

f(a+h)= f(a) +hf'(a) +f(a+

2! 6h).
We put a =45°,h =1°.
Then from above we get

fla+h)= f(a)+ hf (a)

or, f(46°) = f(45° )+

1° cos 45o

180 )
EXERCISE
(L] Short Answer Questions
1. Verify Rolle's

)
theorem in each of the

fo)= X, 0sx<l
following cases :
-0, X=1.
(i) f)=4x -x,0Sxs4.
(ii) f)= (x-1)2/)+2, 0sxs2.
(iv)f(x)=x'
-5x+6

)
in 2sxS3. U.T2001
[WB.
2.Verify Rolle's
theorem for the
functions :
f)=x