Asymptotic Notations - GATE Bits in PDF

Asymptotic Notations is an important chapter in Design and Analysis of Algorithms,

which carries over to bigger topics later on. It is useful for all of Algorithms in GATE CS,
BARC, BSNL, DRDO, ISRO, and other exams. These GATE Bits on Asymptotic Notations
can be downloaded in PDF for your reference any time. Use these GATE Study Notes to
help you ace any exam.
When we compare the relative performance of alternative algorithms to solve same
problem, we do not need exact time or space respectively. (we are permitted to be little
sloppy).
Analysis is meaningful only when the input size is large. Assume two algorithms A1 and
A2 are proposed to solve a particular problem. and
TA1(n) = 100n2 + 106n
TA2(n) = n3
Now at first sight it might appear that A2 is better compared to A1, but for large input
size (n) only the order of growth is relevant.
Hence A1 wins over A2, because we will say time complexity of A1 is order of n2 and
time complexity of A2 is order of n3.
To depict time and space complexities of an algorithm asymptotically (approximately)
we require asymptotic notations.

Asymptotic Notations
i) Big – O (O)

1|Page
After analysing an algorithm A, if one says
TA(n) = O(n2)
He means that algorithm will be completed within cn2 time for a sufficiently large n.
Hence big – O given upper bound but this upper bound may or may not be the tightest.
Hence
If TA = O(n2) then
TA = O(n3)
TA = O(n4)

In fact T = O n

[but it does NOT mean O(n2) = O(n3) = O(n4) …]

But first statement is more meaningful

In fact, it means n2 ϵ O(n3)

Set definition O notation
O(g(n)) = {f(n) ∃ c > 0, n0 > 0 such that 0 ≤ f(n) ≤ cg(n), ∀ n ≥ n0}

Hence set O(g(n)) is the umbrella under which all asymptotically smaller functions will reside

Example for O(n4)

n ϵ O(n4)
n2 ϵ O(n4)
n3 ϵ O(n4) etc..

Also, f(n) = n2 + O(n3)

means
f(n) = n2 + g(n)
where g(n) ϵ O(n3)

2|Page
f(n) = O(g(n))
After n = n0 f(n) will never catch c.g(n) where c is a constant
By writing f(n) = O(g(n)) we mean that a function f(n) is a member of set O(g(n))
Since O notation describes an upper bound, it can be used to bound the worst case
running time of an algorithm.

Hence it is meaningful to say that running time of insertion sort is O(n2), though we can
also say O(n3), because again big –O need not provide tightest upper bound. [It may or
may not be tightest].

ii. Small–o
Big –O notation is like ≤
f(n) ≤ g(n) when f(n) = O(g(n))
small – o notation is like <
f(n) < g(n) when f(n) = o(g(n))
The asymptotic upper bound provided by O – notation may or may not be
asymptotically tight.
Small – o is used to denote an upper bound that is NOT asymptotically tight.
o(g(n)) = {f(n): for any positive constant c > 0, n0 > 0, 0 ≤ f(n) < cg(n) ∀ n>n0}
Example
3n = o(n2), but 2n2 ≠ o(n2)

3|Page
logn = o(n), but log n ≠ o(100 logn)
Therefore, when
f(n) = o(g(n)) and n approaches infinity
( )
→∝ =0
( )

iii. Omega (Ω)

Ω notation is used to provide an asymptotic lower bound.
Ω(g(n)) is the set of functions
Ω(g(n)) = {f(n): ∃ c > 0 and n0 such that 0 ≤ c.g(n) ≤ f(n), for all n ≥n0}

f(n) = Ω(g(n))
As Ω notation describes a lower bound, it is used to bound the best case running time of
an algorithm.
Example:
If T(n)denotes the running time of the insertion sort on input – size n, then we can say
that T(n) = Ω(n) as insertion sort takes linear time in best case.
Remember in practice we use O and Ω notations to tight upper and lower bounds
respectively otherwise for any function T(n)both the following statements are TRUE.
T(n) = Ω(0)
T(n) = O(∞)
Which means any algorithm will take at least zero time and at most infinity time to
compute.

4|Page
iv. Little –Omega (ω)
ω –nation is used to denote a lower bound that is not asymptotically tight.
Hence ω(g(n)) is the set
ω(g(n)) = {f(n): ∃ positive C, n0 > 0 such that 0 ≤ C g(n) < f(n) for all n ≥n0}
Ω-notation is like ≥ where us ω notation is like >
Hence

√ = (lg ) √ ≠ √

v. Theta-Notation (θ)
If two functions f(n)and g(n)are of same order, then
f(n) = θ(g(n))
we say that g(n) is an asymptotically tight bound for f(n)
θ(g(n)) the set of functions θ(g(n)) = {f(n): ∃ c1 , c2 >0
and n0 s.t. 0 ≤ c1 g(n) ≤ f(n) ≤ c2 g(n) for all n ≥ n0}
which means
θ(g(n)) = O(g(n)) ∩ Ω(g(n))

Hence depending upon the value of

c, c g(n) can be the tightest upper and lower bound of f(n)
5|Page
Example:
Assume that in the Best Case, Average case and worst case time required by an
algorithm A is 100n, 50nlogn and 10n2 respectively and assume we represent time
complexity of A as TA(n)
Then All of the following statements are TRUE.
T(n) = O(n2)
T(n) = Ω(n)
T(n) = o(n3)
T(n) = ω(√n)

Example:
As we know that no matter what the input is mergesort always takes c.nlogn time.
Assume T(n) represents T.C. of mergesort then
T(n) = O(nlogn)
T(n) = Ω(nlogn)
Hence
T(n) = θ(nlogn)
Hence depending upon the behaviour of the algorithm on different inputs we use the
most expressive and meaningful notation to represent complexity of an algorithm
Following statement about time complexity of mergesort are also correct but do not
make much sense:
T(n) = O(n2)
T(n) = o(n2)
T(n) = ω(n)
T(n) = Ω(n)

Now Try It Yourself

1. Which of the following set is empty?
(1) ( ) ( )
(2) ( ) Ω ( )

6|Page
(3) ( ) ( )
(4) ( ) ∩Ω ( )

2. Assume ( ) denotes the time complexity of merge sort algorithm which of the
following statements is/are correct?
(i) ( ) = ( log )
(ii) ( ) = ( log )
(iii) ( ) = ( log )
(iv) ( ) = ( )
(1) i and iii only
(2) ii and iii only
(3) i and iv only
(4) ii, iii and iv only

3. Identify the FALSE statement:

(1) ( ) = ( )=

(2) ( ) = ( ) lg ( ) = ( )
( ) ( )
(3) ( ) = ( ) 2 = 2
(4) ( ) + ( ) = ( ), ( )

4. Given f(n) = ω(n2). Which of the following can never hold?

(1) f(n) = O (n3)
(2) f(n) = Ω (n2)
(3) f(n) = θ (n2)
(4) f(n) = ω (n)

5. Which of the following can never hold?

(1) n3 + Ω (n2) = O (n4)
(2) n + θ (n2) = Ω (n)
(3) n2 + O (n2) = θ (n3)
(4) n3 + O(n3) = O (n4)

6. Assume f(n) and g(n) are two functions such that f(n) = O (g(n)). Which of the
following will always hold?
(1) ( ) = ( )

7|Page
(2) ( ) = Ω ( )
(3) ( ) = ( )
(4) ( ) = Ω ( )

7. Which of the following arrangements of functions is in ascending order of growth

rate. That is if g(n) follows f(n) than it should be the case that f(n) is O(g(n)).

(1) log , log(log ), 2 , log

(2) (log ) , ,2 , log

(3) √ , 2 , . (log ) ,

(4) 2 , √ , (log ) ,

(answers & solutions at the end)

Liked this article on Asymptotic Notations? Let us know in the comments. You may also
like...

Theory of Computation

Rank of a Matrix & Its Properties

Eigen Values & Eigen Vectors

Linear Algebra Short Quiz

ANSWERS & SOLUTIONS to TRY IT YOURSELF QUESTIONS

Ans 1: 1
Solution: There does not exist any function ( ) such
that ( ) ( ) and ( ) ( ) at the same time.
Both the sets in option(A) are mutually exclusive.

Ans 2: 4

8|Page
Solution: we know that time complexity of Merge Sort is ( log ) hence all statements
are valid except i, because little-oh notation is used to give non-tightest upper bound.
Statement (iv) is true because of the fact that if merge sort can be performed in nlogn
time than it can definitely be performed in O(n2),this statement may not be very useful
though.
Remember: Upper bound means "cannot be asymptotically worse than". Therefore all
algorithms are O(∞). Lower bound means "cannot be asymptotically better than". So all
algorithms are Ω(0).

Ans 3: 3
. 3 = ( ) 2 ≠ (2 )

Ans 4: 3
Solution: Statement C can never hold because if tight bounds are Ω(n2) and O(n2) then
ω(n2) can never be the non-tightest lower bound.

Ans 5: 3
Solution: Statements A, B and D can hold but Statement C can never hold.

Ans 6: 4
Solution: Assume ( ) = then option ‘A’ will not hold.

Ans 7: 4

9|Page