Lec 2 KNJ

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

Multivariable Calculus

Dr. S.K. Gupta


Department of Mathematics
Indian Institute of Technology, Roorkee

Lecture - 02
Limits for multivariable functions- I

Hello friends. So, welcome to lecture series on multivariable calculus in the last lecture
we have seen that what do we mean by a function of several variables, domain and range
of two or more than two variable functions. Whether, it is open, closed, bounded, un-
bounded. All these things we have seen in the last lecture. Now, how can you find limit
of multivariable functions.

(Refer Slide Time: 00:58)

So, we recall it first for limits of one variable functions, suppose we have
x lim
 it  a f ( x)  L

So, what does that mean; it means that as x approaches to a, f(x) approaches to L so,
roughly speaking when we say that limit x extend to a, f(x) equals to L. This means as f
tending to a, f(x) tends to L, now how can we defined this mathematically. So, the
mathematical definition of limit is; for every >0,  a corresponding real number >0
such that |f(x) - L| < whenever 0 <| x - a|< less .
You take any >0, no matter how small how large epsilon you are taking. There will
always exist a corresponding >0 such that |f(x) - L|<. Now what does this definition
mean, let us see.

(Refer Slide Time: 03:23)

Now here we are having x lim


 it  a f ( x)  L ;this means for every >0 there exist
corresponding >0 such that |f(x) - L| < whenever 0 < |x - a|< .

Now, this means that x - a < and x - a >- and this means that x <+ and x >- , and of
course, x should not equal to a because if x equal to a then it is equal to 0 and so x not
equal to a. This we also call as a deleted neighbourhood of x at a.

Now, similarly this inequality represent: f(x) - L < and f(x) - L >- g or f(x) < L+ and
f(x) > L-.

Now, this definition means you take any  > 0, no matter how small how large  you are
taking there will always exist a  such that the image of all those x lying in this interval
will always be contained in this band.
Now, if you take  very small tending to 0 then this delta will definitely tending to zero;
that means, that as x approaches to a; f(x) will approach to L because as  tend 0, f(x)
will tends to L and as  tend to 0, x will tends to a. So, x extend to a; f(x) will tends to L.
So, we can say that f of all those points  (-,), will be contained in (L-, L+); So that
means, for every  > 0 there will exist a deleted neighbourhood of x at a, such that the
image of all those x lying in this interval will be contained totally inside this band, totally
inside this interval. So, this is how we can defined limit of single variable functions.

(Refer Slide Time: 07:55)

Now, let us first discuss few examples based on this for example, x lim
  it
 x (2 x  1)  5 .

This is very simple you can easily see that limit of this simple is 5; how can we prove
this using delta epsilon definition.
So, let us see how can we prove, let  > 0 be given, you take any  > 0 . So we can say
that for  > 0 ,  some . So, basically what we have to show, we have to show that for
any  > 0 there will always exist corresponding  such that definition holds. What we
have to prove basically that for every  > 0 , there will exist corresponding  > 0, such
that |2x+1-5|<, whenever 0 < |x - 1|<.

So, basically x - 2 < ; so, basically we have to show the correspondence of  in terms of
. We have to show the existence of . So, how can we do that. Now, we start with this
inequality this is|2x+1-5|, this is equals |2x-4|, this is equal to 2|x-2|, and this is equals to
, if you take this is less than delta. So, this is less than delta. Now this quantity; 2|x-2|
<2; because |x-2|<. Now, if we choose 2 ≤, then| 2 x +1 - 5| <  whenever

0<| x-2|<.

Now, this is because whenever you choose ≤/2; then this inequality always holds
because this is less than 2 delta which is less than equal to epsilon so; that means, this is
less than this is less than epsilon whenever this is less than delta. So, this inequality
always holds whenever delta is less than equal to epsilon by 2 you choose any delta
greater than 0 you can always you can choose any epsilon greater than 0, you can always
find delta which is less than equal to epsilon by 2 for which this inequality holds.

So, we have shown the existence of  in terms of ; for different ,  will be different, but
we have shown the existence of  in terms of  such that this inequality holds. Hence, we
c can say that this limit exists and is equal to 5.

So, this is how using delta epsilon definition we can show the existence of a limit of a
function. Now, same concept we extend for 2 variable functions, how can you do that let
us see, for 2 variable functions we say that f approaches the limit L as (x,y) approaches
to (x0,y0)
(Refer Slide Time: 12:11)

If the values of f(x, y) lie arbitrarily close to a fixed real numbers L for every (x, y)
sufficiently close to (x0, y0). You take any (x, y) that is sufficiently close to (x 0, y0), f(x)
will be arbitrarily close to L; that means, f(x) will approach to L as (x,y) approaches to
(x0, y0).

How can you define this in terms of delta epsilon let us see, Now here we are having two
variable functions instead of a single variable function.

(Refer Slide Time: 12:57)


Now, in a single variable functions say we have x = a; so, when we take disk at this point
at x = a. So, this will be interval, it is a - to + , this is an interval. Now, when we take
point in two variables; a function of two variable say here (x0, y0). Now instead of a
interval it will be a disk centred at this point, it will be a disk centred at (x 0, y0). So, how
can I define limit now, limit (x, y)  (x0, y0) = L. So, how can we prove whether limit
exists and is equal to L.

So, for that again we will repeat the same definition for a two variable functions for all
> 0,  corresponding the real number >0 such that |f(x, y) - L <, whenever

0  ( x  x0 ) 2  ( y  y 0 ) 2   So, you take any >0 there will always exist a corresponding
real number >0 such that this inequality hold whenever this is less than  or greater than
zero. So, this is a deleted neighbourhood of (x 0, y0) of radius  so, this is a basically a
circle radius .

Now, again what does it mean; it means that; this inequality means that f(x, y) is less
than L +  to L -  and this means all (x, y) data lies inside the region inside a disk of
radius  and centre (x0, y0);

Now, you choose any > 0 there will always exist a corresponding  such that all those
(x, y) which lie in this disk, now image of all those (x, y) is lying in this disk. No matter
how small  or how large  you are taking, there will always exist corresponding  such
that the image of all those (x, y) lying in this disk will totally contain in this band.

So, this definition can also be written as |f(x, y)-L|< whenever 0<| x - x0|< and
0<|y - y0|<; that means, instead of disk it may be a rectangle, then also we can apply the
same definition. Now, let us discuss few examples based on this so, that concept of limit
will be more clear.
(Refer Slide Time: 18:02)

First we have some properties of limit, suppose limit (x, y)(x0, y0), f(x, y) is L and limit
(x, y)(x0, y0), g(x, y) is M, then the addition and subtraction of f(x, y) and g (x, y) limit
(x, y)(x0, y0) is L ± M. And similarly we have the product of two functions then the limit
will also be product. Then scalar multiplication will be K into L, division f/g will be
simply L/M. M should not equal to 0 and similarly we have the next property. These are
very straightforward.

Now, come to problems based on delta epsilon definition.

(Refer Slide Time: 18:53)

The first problem it is limit (x, y) ( 1, 2), 2 x +y = 4.


(Refer Slide Time: 18:59)

So, how can we prove it, when you substitute x = 1 and y = 2, the value is 2 + 2 which is
4. Now, if somebody asked how can we prove mathematically that this limit exists and is
equal to 4. So, we have the only option is delta epsilon definition. We can use delta
epsilon definition to show that this limit exists and equal to 4, how can you proceed for
that, let  > 0 be given.

Now, again we have to show the correspondence of a  corresponding  > 0; such that
we have to show that there exists a corresponding  > 0 such that |2 x + y |< 4< 

whenever 0 ( x  1) 2  ( y  2) 2   or |2x + y - 4|< , whenever 0<|x- 1 |<  and 0 <|


y - 2 |< .

So, we can use any definition either this or this to prove this result now you take |2 x+ y-
4|=|2x - 1 + y - 2|. You can easily see that it is 2x+y which is 2x+y; - 2 - 2 is - 4. Now,
this is less than or equals to|2 x- 2 x - 1| + | y - 2| because | a+b| ≤ | a | + | b |.

Now, this is equal to 2 | x - 1 + y - 2| . so, it will be better if you apply this definition
because this is direct for this particular problem. Now, |x-1| <  , if you take so, this is
less than 2 and this is again  which is 3. So, if we choose or if you take 3 ≤ then |
2x+5-4|<, whenever 0<|x-1|<  and 0 <|y-2| <. If we choose 3 ≤, then this quantity
will be less than , whenever this is less than 
So, this inequality holds if this ≤/3. So, for any  we have shown the corresponding 
existence of corresponding  such that this inequality holds hence we can say that this
limit exists and is equal to 4.

(Refer Slide Time: 23:05)

So, the next example, Now let us discuss this is simple, now here it is basically a 0/0
form so, it is very difficult to say whether this limit exists or not. Again if this limit exists
so, we have to show this by delta epsilon definition; that means, we have to show the
existence of  corresponding to every > 0. So, how can you proceed. Supposes limit
exist and we have to show the existence of this limit

xy xy
So, let > 0 be given now you take 2 2 , this is equals to | |, Now, we know
x y x  y2
2

that (x-y)2 ≥0 so, x2+y2-2xy≥0. So, x y ≤ (x2 + y2)/2 so, mod of this quantity will also be
less than or equals to x2+ y2

x2  y2 x2  y2
So, this is less than or equals to , this is equals to Now, what we
2 x2  y2 2

xy
have to show here we have to show that| -0|<, whenever 0 < x2  y2 <.
x  y2
2
xy
So, this quantity will be less than /2 so, if we choose /2 ≤, then| | <
x  y2
2

whenever 0 < x2  y2 <.. So, if you take any  satisfying this inequality, this inequality
will be satisfied. So, we have shown that this term to  corresponding to any  such that
this inequality holds. So, hence we can say that this limit exists and is equal to 0 so, in
this way we can prove that using delta epsilon definition that the limit of two or more
than two variable function exists and is equal to L. Now, we will see some more
properties of limit of several variable functions in the next lecture.

Thank you very much.

You might also like