Professional Documents
Culture Documents
&
QUEUEING THEORY
Dr. P. GodhandaRaman
M.Sc., M.A., M.B.A., Ph.D.
Assistant Professor (Senior Grade)
Department of Mathematics
Faculty of Engineering & Technology
SRM Institute of Science and Technology
Kattankulathur – 603 203
Email : godhanda@gmail.com, Mobile : 9941740168
1
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
15MA207 – PROBABILITY AND QUEUEING THEORY
UNIT – II - THEORETICAL DISTRIBUTIONS
Syllabus
• Binomial Distribution
• Poisson Distribution
• Geometric Distribution
• Uniform Distribution
• Exponential Distribution
• Normal Distribution
Theoretical
Binomial = = , = 0,1, … +
! "#
Poisson = = , = 0,1, … ∞
!
1
= = "
, = 1, … ∞ %
1−
Geometric
CONTINOUS DISTRIBUTION
Moment Generating
Mean Variance
Continuous Probability Density Function (p.d.f.) Function (m.g.f.)
1 *
− + '+( '−( %
& = , (< <'
'−( , '−( 2 12
Uniform
1 1
& = , ≥0
−,
Exponential %
2 567 3
" 4 9
& = 3 8 , => ?
!3 83
@ : ;%
Normal / √%1 3
−∞ < < ∞, −∞ < : < ∞, ; > 0
BINOMIAL DISTRIBUTION
Bernoulli Distribution : A Bernoulli distribution is one having the following properties
(i) The experiment consists of repeated trials.
(ii) Each trial results in an outcome that may be classified under two mutually exclusive categories as a success or
as a failure.
(iii) The probability of success denoted by , remains constant from trial to trial.
(iv) The repeated trials are independent.
Binomial Distribution:
A Bernoulli trial can result in a success with probability and a failure with probability = 1 − . Then the
probability distribution of the binomial random variable , the number of successes in independent trials
A = B = CDB EB FC B , B = G, H, … C
The quantities & are called the parameters of binomial distribution.
2
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Areas of Application:
1. Quality control measures and Sampling processes in industries to classify items as defective or non defective.
2. Medical applications as success or failure of a surgery, cure or no cure of a patient.
Moment Generating Function (m.g.f.) in Binomial Distribution
IJ , = ∑ LM = ∑ LM = ∑ LM
= M M M
+ " " "
+ % % %
+ ⋯+
= + " " "
+ % % %
+ ⋯+ = F + EO C
Mean and Variance using Moment Generating Function in Binomial Distribution
R R
P = Q IJ , S =Q + S =T + "
U LM = + M "
M
= CE
R LM R LM
R3 R
P %
=Q I , S =Q + "
S = T −1 + %
+ + " U
R 3 J LM R LM
LM
= T −1 + M %
M
+ + M " MU
= T − 1 + 1U = CV EV − CEV + CE
W %
=P − TP U% =
% %
− %+ − % %=− %+ = 1 − = CEF
Problems in Binomial Distribution
1. Four coins are tossed simultaneously. What is the probability of getting 2 heads and at least 2 heads?
" " "
Solution : = 4, = , = 1 − =1− = , = = , = 0,1, … .
% % %
" % " \ % ]
(i) 2 ℎ (Z[ = =2 =4 % 4% 9 4% 9 =^
(ii) (,_ ([, 2 ℎ (Z[ = ≥2 = =2 + =3 + =4
" % " \ % " ] " \ ] " \ " \ \ ] " " ""
=4 % 4%9 4%9 +4 ] 4%9 4%9 +4 \ 4%9 4%9 = + + =
^ \ "a "a
2. The probability that a patient recovers from a disease is 0.3. If 18 people are affected from this disease,
What is the probability that (i) At least 10 survive (ii) Exactly 6 survive (iii) 4 to 7 survive
Solution: = 18, = 0.3, = 1 − = 1 − 0.3 = 0.7, = = 18 0.3 0.7 "^
(i) ≥ 10 = 1 − < 10 = 1 − T 0 + 1 + ⋯ + 9 U = 1 − 0.9790 = 0.021
(ii) = 6 = 18 a 0.3 0.7 "^ a = 0.1873
a
(iii) 4 ,g 7 [hijkl = =4 + =5 + =6 + =7
= 18 \ 0.3 0.7\ "^ \
+ 18 n 0.3 0.7
n "^ n
+ 18 a 0.3 0.7 "^ a + 18 o 0.3 o 0.7 "^ o = 0.6947
a
3. In a large consignment of electric bulbs 10% are defective. A random sample of 20 is taken for inspection. Find
the probability that (i) All are good (ii) At most there are 3 defective (iii) Exactly there are 3 defective bulbs.
"M
Solution : = 20, = 10% = "MM = 0.1, = 1 − = 1 − 0.1 = 0.9, = = 20 0.1 0.9 %M
3
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
" " \
Let X denote the no. of successes (boys). = = , = 0,1, … . = =4 4% 9 4% 9
" % " \ % ]
(i) 2 'gt[ ( Z 2 uki_[ = =2 =4 % 4%9 4%9 =
^
]
Number of families having 2 boys and 2 girls = r = 2 = 800 4 9 = 300
^
" M " \ M " "n
(ii) At least 1 boy = ≥1 =1− =0 =1−4 M 4% 9 4% 9 =1− =
"a "a
"n
Number of families having At least 1 boy = r ≥1 = 800 4 9 = 750
"a
(iii) At most 2 girls = exactly 0 , 1 or 2 girls = 4 + 3 + 2 =1−T 0 + 1 U
" M " \ M " " " \ " ""
=1−4 M 4%9 4%9 −4 "4 9 4 9 =
% % "a
""
Number of families having At most 2 girls = 800 4"a9 = 550
(iv) Children of both sexes = 1 − children of the same sex = 1 − T all are boys + all are girls U
" \ " \ \ " M " \ M o
=1− 4 − 0 =1−4 \ 4% 9 4% 9 −4 M 4% 9 4% 9 =
^
o
Number of families having Children of both sexes = 800 4 9 = 700
^
6. For a binomial distribution the mean is 6 and variance is 2. Find the distribution and find A = H .
‹Œ % H " V
Solution : Mean = = 6, Wariance = =2, ‹ =a⇒F=Ž, =1− =1−]=Ž
a a "^ V H % " " ‘ "
= 6 ⇒ = ⇒ = 3 = = 9, C = •, E = , F = , =1 =9 " 4] 9 4] 9 = 0.0012
‹ 4 9 % Ž Ž
•
7. A Binomial variable satisfies the relation •A = ’ = A = V when C = “. Find the parameter E. Solution
: = 6, = = , = 0,1, …
9 =4 = = 2 ⇒ 9 × 6 \ \ a \ = 6 % % a % ⇒ 9 × 6 \ \ % = 6 % % \
9 × 6 \ % = 6 % % ⇒ 135 % = 15 % ⇒ 9 % − % = 0 ⇒ 9 % − 1 − % = 0 ⇒ 8 % + 2 − 1 = 0
" " H
= − gi , E= ∵ E •–CC— ˜O CO™– š›O
% \ ’
H Ž œ
8. A discrete RV X has moment generating function = 4 + O 9 . Find , –• –Cž A =V .
’ ’
] " "n "n
Solution : IJ , = + , = \
, = \
, = 5, P = = \
, W = = "a
] % " n % \n
=2 =5 % 4\ 9 4\ 9 =
n"%
9. Fit a binomial distribution for the following data. Find the parameters of the distribution.
B 0 1 2 3 4 5 6 Total
Ÿ 5 18 28 12 7 6 4 80
Solution : Fitting a binomial distribution means assuming that the given distribution is approximately binomial and
hence finding the probability mass function and the finding the theoretical frequencies.
To find the binomial frequency distribution r + , which fits the given data, we require r, ( Z .
We assume r = total frequency = 80 and = no. of trials = 6 from the given data.
0 1 2 3 4 5 6 Total
& 5 18 28 12 7 6 4 80
& 0 18 56 36 28 30 24 192
∑¡ "‘%
̅= ∑¡
= = 2.4 , = 2.4 ⇒ 6 = 2.4 , = 0.4, = 1 − = 0.6
^M
Theoretical frequencies are given by N = = r , = 0,1, …
80 = 0 = 80 × 6 M 0.4 M 0.6 a M = 3.73, 80 = 1 = 80 × 6 " 0.4 " 0.6 a " = 14.93
80 = 2 = 80 × 6 % 0.4 0.6 % a %
= 24.88, 80 = 3 = 80 × 6 ] 0.4 ] 0.6 a ] = 22.12
80 = 4 = 80 × 6 \ 0.4 \ 0.6 a \ = 11.06, 80 = 5 = 80 × 6 n 0.4 n 0.6 a n = 2.95
80 = 6 = 80 × 6 a 0.4 0.6 a a a
= 0.33, Converting these values into whole numbers consistent
with the condition that the total frequency is 80, the corresponding binomial frequency distribution is as follows
X 0 1 2 3 4 5 6 Total
Theoretical f 4 15 25 22 11 3 0 80
4
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
POISSON DISTRIBUTION
Definition : The probability distribution of the Poisson random variable , representing the number of outcomes
O6¢ ¢B
occurring in a given time interval or specified region represented as ,, is A =B = , B = G, H, … ∞. where
B!
is the average number of outcomes per unit time or region.
Poisson distribution as Limiting Form of Binomial Distribution
Poisson distribution is a limiting case of binomial distribution under the following conditions
(i) , the number of trials is indefinitely large, i.e., → ∞.
(ii) , the constant probability of success in each trial is very small, i.e., → 0.
(iii) = is finite or = and = 1 − , where is a positive real number.
Areas of Application:
1. The number of misprints on a page of a book.
2. The number of deaths due to accidents in a month on national highway 47.
3. The number of break downs of a printing machine in a day.
4. The number of vacancies occurring during a year in a particular department.
Moment Generating Function (m.g.f.) in Poisson Distribution
5
6¥ 5 5 !# B BV
IJ , = ∑¤LM = ∑¤LM = ∑¤LM = ∑¤LM 4∵ OB = H + H! + +⋯9
! ! ! V!
! #§ ! #2 ! #3 ! #2 ! #3
= ¦ + + + ⋯¨ = ¦1 + + +⋯¨ = !
= O¢ O H#
M! "! %! "! %!
Mean and Variance Using Moment Generating Function in Poisson Distribution
R R ! ! §
P = Q IJ , S =Q "#
S =© "#
ª LM
= "#
M
= ¢ ∵ OG = H#
R LM R LM
R3 R
P =Q I , S =Q = © + ª = ¢ ¢+H
! ! !
%
R 3 J R
"#
S "# "#
LM
LM LM
W =P %
− TP
+ − %=¢ U% = %
GEOMETRIC DISTRIBUTION
Definition: If repeated independent trials can result in a success with probability and a failure with probability
= 1 − , then the probability distribution of the random variable , the number of trials on which the first success
occurs, is A = B = E FB H , B = H, V, … ∞ .
Application: Geometric distribution has important application in queueing theory, related to the number of units
which are being served or waiting to be served at any given time.
Moment Generating Function (M.G.F.) in Geometric Distribution
‹ ‹
IJ , = ∑¤LM = ∑¤L" "
= ∑¤L" = T " + % + ] + ⋯U
Œ Œ
‹ EO
= T1 + "
+ %
+⋯U = T1 − U "
=H ©∵ H − B H
= H + B + BV + ⋯ ª
Œ F O
6
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Mean and Variance using Moment Generating Function in Geometric Distribution
R ‹ ! R " Œ ! #‹ ! ‹ ! Œ ! # ž µ ›µ¶ µ›¶
P = Q IJ , S =Q 4 9S =Q S Q∵ žB 4›9 = S
R LM R "
Œ !
LM " Œ ! 3 LM ›V
‹ ! ‹ Œ 3! ?‹ Œ 3! ‹ ! ‹ § ‹ H
= Q " Œ ! 3
S = Q " Œ ! 3 S = " Œ § 3 = ‹3 = E ∵E+F=H
LM LM
3 2
R3 R ‹ ! " Œ ! # ‹ ! ‹ ! % " Œ ! # Œ ! # ž
P %
=Q I , S =Q S =¦ ¨ Q∵ žB µ › = µ ›· + › µ· S
R 3 J LM R " Œ ! 3 LM " Œ ! ®
LM
3 2
" Œ § # ‹ § ‹ § % " Œ § # Œ § # ‹• ?%‹3 Œ % ‹?%Œ H?F
=¦ ¨= = 4 ‹® 9 =
" Œ § ® ‹® EV
"?Œ " "?Œ " F
W =P %
− TP U =
%
− = =
‹3 ‹3 ‹3 EV
ANOTHER FORM OF GEOMETRIC DISTRIBUTION
Definition: If X denotes the number of failure before the first success, then A = B = E FB, B = G, H, V, … ∞ .
Moment Generating Function (M.G.F.) in Geometric Distribution
E
IJ , = ∑¤LM = ∑¤LM = T1 + " + % + ] + ⋯ U = T1 − U " =
H F O
Mean and Variance using Moment Generating Function in Geometric Distribution
R R ‹ R
P = Q IJ , S =Q 4 9S =Q 1− "
S = T −1 1 − " "
− U LM
R LM R " Œ ! LM R LM
F
=T 1− %U
LM = M
1− M %
= ∵E+F=H
E
3 2
R3 R ‹ Œ ! " Œ ! # ‹ Œ ! ‹ Œ ! % " Œ ! # Œ ! # ž µ ›µ¶ µ›¶
P %
=Q I , S =Q S =¦ ¨ Q∵ žB 4›9 = S
R 3 J LM R " Œ ! 3 LM " Œ ! ® ›V
LM
" Œ § #3 ‹ Œ § ‹ Œ § % " Œ § #2 Œ § # ‹• Œ?%‹3 Œ 3 ‹?%Œ FV ?F
P %
=¦ ¨= = %
4 9=
" Œ § ® ‹® ‹® EV
Œ 3 ?Œ Œ3 Œ 3 ?Œ Œ3 F
W =P %
− TP U% = − 3 = =
‹3 ‹ ‹3 EV
Memory less Property of Geometric Distribution
If is a RV with geormetic distribution, then lacks memory, in the sense that A >¸+ ⁄ >[ =A >, .
Proof : = = "
, = 1, … ∞
º J»¼? ∩ J»¼ º J»¼?
> [ + ,⁄ > [ = º J»¼
= º J»¼
>¾ = ∑¤L¿?" "
= ¿?" " + ¿?% "
+ ¿?] "
+⋯= ¿
+ ¿?"
+ ¿?%
+⋯
‹ Œ À
= ¿
1+ "
+ %
+⋯ = ¿
1− "
= = FÁ
‹
Œ ÂÃ! Œ  Œ !
Hence > [ + , = ¼?
and >[ = ¼
; > [ + ,⁄ > [ = = = =A >,
ŒÂ ŒÂ
Problems in Geometric Distribution
1. If the probability that an applicant for a drivers license will pass the road test on any given trial is 0.8. What is the
probability that he will finally pass the test (i) On the fourth trial (ii) In less than 4 trials?
Solution : = 0.8, = 1 − = 0.2, = = "
, = 1, … ∞
(i) = 4 = 0.8 0.2 \ "
= 0.0064
(ii) < 4 = 1 + 2 + 3 = 0.8 T 0.2 " " + 0.2 % " + 0.2 ] " U = 0.992
2. A typist types 2 letters erroneously for every 100 letters. What is the probability that the 10th letter typed is the 1st
erroneous letter?
%
Solution : = = 0.02, =1− = 0.98, = 10 = 0.02 0.98 "M "
= 0.0167
"MM
3. A die is tossed until 6 appears. What is the probability that it must be tossed more than 5 times?
" n
Solution : = , = 1 − = , = = "
, = 1, … ∞
a a
> 5 = 1 − ≤ 5 =1−T 1 + 2 + 3 + 4 + 5 U
" n " " n % " n ] " n \ " n n "
=1 − a ¦4a9 + 4a9 + 4a9 + 4a 9 + 4a 9 ¨ = 0.4019
4. A trainee soldier shoots a target in an independent fashion. If the probability that the target is shot on any one
shot is 0.8. (i) What is the probability that the target would be first hit at the 6th attempt? (ii) What is the
probability that it takes less than 5 shots?
7
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Solution : = 0.8, = 1 − = 0.2, i = 6 = 0.8 0.2 a " = 0.00026
(ii) < 5 = 1 + 2 + 3 + 4 = 0.8 T 0.2 " " + 0.2 % " + 0.2 ] " + 0.2 \ " U = 0.9984
5. The probability that a candidate can pass in an exam is 0.6. (i) What is the probability that he pass in the 3rd trial
(ii) What is the probability that he pass before the 3rd trial?
Solution : : = 0.6, = 1 − = 0.4 i = 3 = 0.6 0.4 ] " = 0.096
(ii) < 3 = 1 + 2 = 0.6 T 0.4 " " + 0.4 % " U = 0.84
6. A discrete RV X has moment generating function = œ − ’ O H find A = œ —• “ .
Solution : = = , = 0, 1, … ∞, IJ , = 1 − "
8
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
" "
Solution: & = , (< < ', & = , 0< < 30
* + ]M
"n " ]M " " " "
(i) < 5 Äk h, [ = 10 < < 15 + 25 < < 30 = Ê"M Z + Ê%n Z = T U"n
"M + ]M T U%n = ]
]M
]M ]M ]M
] " "^ " " " "
(ii) Ï,_ ([, 12 Äk = 0< <3 + 15 < < 18 = ÊM ]M Z + Ê"n Z = T U]M + T U"^ "n = n
]M ]M ]M
4. Trains arrive at a station at 15 minutes intervals starting at 4 am. If a passenger arrive to the station at a time that
is uniformly distributed between 9.00 and 9.30, find the probability that he has to wait for the train for
(i) Less than 6 minutes (ii) More than 10 minutes.
" "
Solution: & =* +
, (< < ', & = ]M , 0 < < 30
"n " ]M " " " %
(i) < 6 Äk h, [ = 9< < 15 + 24 < < 30 = Ê‘ Z + Ê%\ Z = T U"n ‘ + ]M T U%\ = n
]M
]M ]M ]M
n " %M " " " "
(ii) Igi ,ℎ( 10 Äk = 0< <5 + 15 < < 20 = ÊM Z
]M
+ Ê"n ]M Z = ]M T UnM + ]M T U"n
%M
=
]
5. If X is a RV with a continuous distribution function Ð B , prove that Ñ = Ð has a uniform distribution in
H
B − H , H ≤ B ≤ ŽÉ
G, H . Further if Ÿ B = ÆV find the range of Ñ corresponding to the range H. H ≤ ≤ V. •.
G, — ÇO•Èš¸O
Solution: The distribution function of Y :ÒÓ t = Ô≤t = Õ ≤t = T ≤Õ "
t U = ÕTÕ "
t U=t,
R
u t = Ò t = 1, Also the range of Y is 0 ≤ y ≤ 1, since the range of F x is (0, 1).
RÖ Ó
EXPONENTIAL DISTRIBUTION
Definition: A continuous RV X defined in 0, ∞ is said to follow an exponential distribution if the probability
density function is Ÿ B = ¢ O ¢B , B ≥ G.
Application: Exponential distribution is useful in queueing theory and reliability theory. Time to failure of a
component and time between arrivals can be modeled using exponential distribution.
Moment Generating Function (M.G.F.) in Exponential Distribution
6 ¥6! 5 ¤
¤ ¤ ¤ ¢
IJ , = Ê ¤
& Z = ÊM Z = ÊM Z = ¦ ¨ =¢
M
Mean and Variance using Moment Generating Function in Exponential Distribution
R R R H
P = Q IJ , S = Q 4 9S = Q − , "S = T −1 − , % −1 U LM =
R LM R LM R LM ¢
R3 R V
P %
= QR 3 IJ , S = QR −, %
S = T −2 −, ]
−1 U LM = ¢V
LM LM
% " H
W =P %
− TP U% = 3 − 3 = ¢V
Memory less Property of Exponential Distribution
Statement: If is exponential distributed with parameter ¢, then for any 2 positive integers s and t
A >¸+ ⁄ >[ =A >, .
º J»¼? ∩ J»¼ º J»¼?
Proof : > [ + ,⁄ > [ = = , & = , ≥0
º J»¼ º J»¼
6¥5 ¤
¤ ¤
> ¾ = Ê¿ & Z = Ê¿ Z = Q S =O ¢Á
¿
6¥ ÂÃ!
Hence > [ + , = ¼?
and >[ = ¼
; > [ + ,⁄ > [ = 6¥Â = =A >,
Problems in Exponential Distribution
H
1. The time (in hours) required to repairs a machine is exponential, distributed with parameter ¢ = . (i) What is the
V
probability that the repair time exceeds 2 hours? (ii) What is the conditional prob. that a repair takes at least 10h
given that its duration exceeds 9h?
5 ¤
5 6 6Ú 62
¤" " 3
Solution: & = , > 0, (i) > 2 = Ê% 3 Z = Ø 2 Ù = = "
= 0.3679
% % "
3 %
5 ¤ 2
5 6 6Ú 6 2
¤ ¤" "
≥ 10⁄ > 9 =
3 3
(ii) >1 = Ê" & Z = Ê" % 3 Z = Ø 2Ù = = 3 = 0.6065
% "
3 "
2. The mileage which car owners get with a certain kind of radial tire is a RV having an exponential distribution
with mean 40,000 km. Find the prob. that one of these tires will last (i) Atleast 20,000 km (ii) At most 30,000 km
5
" " "
Solution: I ( = = 40,000 ¾Ä, = , & = , > 0, & = ®§,§§§ , >0
\M,MMM \M,MMM
9
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
5 ¤
5 6 2
¤ ¤ " " ®§,§§§
(i) ≥ 20,000 = Ê%M,MMM & Z = Ê%M,MMM \M,MMM ®§,§§§ Z = Ø 2 Ù = 3 = 0.6065
\M,MMM
®§,§§§ %M,MMM
5 ]M,MMM
5 6 •
]M,MMM ]M,MMM " " ®§,§§§
(ii) ≤ 30,000 = ÊM & Z = ÊM ®§,§§§ Z = Ø 2 Ù =1− ® = 0.527
\M,MMM \M,MMM
®§,§§§ M
3. The length of time a person speaks over phone follows exponential distribution with mean 6. What is the
probability that the person will talk for (i) More than 8 min (ii) Less than 4 min (iii) Between 4 and 8 min
5 ¤
5 6 ®
" " ¤ " " °
Solution: I ( = = 6, = , & = , > 0 (i) >8 = Ê^ a ° Z = Ø 2 Ù = • = 0.2635
a a
° ^
5 \
5 6 3
\ " " °
(ii) <4 = ÊM a ° Z = Ø 2 Ù =1− • = 0.4865
a
° M
5 ^
5 6 3 ®
^ " " °
(iii) 4≤ ≤8 = Ê\ a ° Z = a
Ø 2 Ù = • − • = 0.5134 − 0.2635 = 0.2499
° \
4. The amount of time that a watch can run without having to be reset is a random variable having exponential
distribution, with mean 120 days. Find the prob. that such a watch will have to be reset in less than 24 days.
5
" " "
Solution:, I ( = = 120 Z(t[, = "%M , & = , > 0, & = "%M 23§ , >0
5 %\
5 6 2
%\ %\ " " 23§
< 24 = ÊM & Z = ÊM 23§ Z = Ø 2 Ù =1− ¯ = 0.1813
"%M "%M
23§ M
5. The number of kilo meters that a car can run before its battery has to be replaced is exponentially distributed with
an average of 10,000 kms. If the owner desires to take a tour consisting of 8000 kms, what is the probability that
he will be able to complete is his tour with our replacing the battery?
5
" " "
Solution:, I ( = = 10,000, = , & = , > 0, & = 2§,§§§ , >0
"M,MMM "M,MMM
5 ¤
5 6 ®
¤ ¤ " " 2§,§§§
> 8000 = Ê^MMM & Z = Ê^MMM 2§,§§§ Z = Ø 2 Ù = ¯ = 0.4493
"M,MMM "M,MMM
2§,§§§ ^MMM
6. In a construction site, 3 lorries unload materials per hour, on an average. What is the probability that the time
between arrival of successive lorries will be (i) at least 30 minutes (ii) less than 10 minutes?
Solution: = 3, & = , > 0, & = 3 ] , > 0
"
(i) Probability that the time between arrival of successive lorries equal to 30 minutes or hour
%
6•5 ¤ •
" ¤ ¤
4 ≥ %
9 = Ê2 & Z = Ê2 3 ]
Z = 3Q S
] 2
= 3 = 0.223
3 3 3
"
(ii) Probability that the time between arrival of successive lorries equal to 10 minutes or hour
a
2
2 2 6•5 ° 2
"
4 < a9 = ÊM° & Z = ÊM° 3 ]
Z = 3Q
] M
S =1− 3 = 0.393
NORMAL DISTRIBUTION
The Normal distribution was first described by De Moive in 1933 as the limiting form of Binomial distribution as the
number of trials becomes infinite. This discovery came into limelight after its discovery by both Laplace and Gauss
half a century later. So this distribution is also called Gaussion distribution.
Definition: A continuous RV X, with parameters : and ; % is normal if it has a probability density function
H B6Ý V
H 4 9
Ÿ B = O V Û , −∞ < <∞
Û√V Ü
J >
Standard Normal distribution: If X is a RV following normal distribution with parameter : and ;, then Þ = is
/
àV
H
called a Standard Normal variate and the p.d.f. of the standard variate Z is given by ß à = O V , −∞ < Þ < ∞
√V Ü
Application: (i) The most important continuous probability distribution in the statistics field is Normal distribution.
(ii) In nature like rainfall and meteorological studies (iii) In industry (iv) In error calculation of experiments
(v) Statistical quality control (vi) Radar applications and in research.
10
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Problems in Normal Distribution
1. If is normally distributed and the mean is 12 and the SD is 4. Find out the following
(i) A ≥ VG (ii) A ≤ VG (iii) A G ≤ ≤ HV
Solution: : = 12, ; = 4
J > %M > %M "%
(i) ≥ 20 = 4/
≥ /
9 = 4á ≥ \
9 = á ≥ 2 = 0.5 − 0 ≤ á ≤ 2 = 0.5 − 0.4772 = 0.0228
J > %M > %M "%
(ii) ≤ 20 = 4 ≤ 9= 4á ≤ 9= á ≤ 2 = 0.5 + 0 ≤ á ≤ 2 = 0.5 + 0.4772 = 0.9772
/ / \
M > J > "% > "% "% "%
(iii) 0≤ ≤ 12 = 4 ≤ ≤ 9= 4 ≤á≤ 9= −3 ≤ á ≤ 0 = 0 ≤ á ≤ 3 = 0.4987
/ / / \ \
2. The savings bank account of a customer showed an average balance of Rs. 150 and a S.D. of Rs. 50. Assuming
that the account balances are normally distributed (i) What percentage of account is over Rs. 200? (ii)
What percentage of account is between Rs. 120 & Rs. 170? (iii) What % of account is less than Rs. 75?
Solution: : = 150, ; = 50
%MM "nM
(i) ≥ 200 = 4á ≥ nM
9 = á ≥ 1 = 0.5 − 0 ≤ á ≤ 1 = 0.5 − 0.3413 = 0.1587
Percentage of account is over Rs. 200 is 15.87%
"%M "nM "oM "nM
(ii) 120 < < 170 = 4 nM
<á< nM
9 = −0.6 < á < 0.4
= 0 < á < 0.6 + 0 < á < 0.4 = 0.2257 + 0.1554 = 0.3811
Percentage of account is between Rs. 120 & Rs. 170 is 38.11%
on "nM
(iii) < 75 = 4á < 9 = á < −1.5 = 0.5 − 0 < á < 1.5 = 0.5 − 0.4332 = 0.0668
nM
Percentage of account is less than Rs. 75 is 6.68%
3. In an examination the marks obtained by the students in Maths, Physics and Chemistry are normally distributed
about mean 50, 52, 48 and S.D. 15, 12, 16 respectively. Find the prob. of securing a total mark of 180 or above.
Solution: Let , Ô, á be the marks of respective subjects. The total marks â = + Ô + á
: = P â = P +Ô+á = P + P Ô + P á = 50 + 52 + 48 = 150
;% = W â = W + Ô + á = W + W Ô + W á = 15% + 12% + 16% = 225 + 144 + 256 = 625 , ; = 25
"^M "nM
â ≥ 180 = 4á ≥ 9= á ≥ 1.2 = 0.5 − 0 ≤ á ≤ 1.2 = 0.5 − 0.3849 = 0.1151
%n
4. In an engineering examination, a student is considered to have failed, secured second class, first class and
distinction, according as he scores less than 45%, between 45% and 60%, between 60% and 75% and above 75%
respectively. In a particular year 10% of the students failed in the examination and 5% of the students got
distinction. Find the percentages of students who have got first class and second class.
Solution: Let X follow the distribution N μ, σ .
Given: < 45 = 0.10 and > 75 = 0.05
J > \n > on > J >
4−∞ < /
<
/
9 = 0.1 and 4 /
<
/
< ∞9 = 0.05
\n > on >
4−∞ < á < / 9 = 0.1 and 4 / < á < ∞9 = 0.05
> \n on >
40 < á < /
9 = 0.4 and 40 < á < /
9 = 0.45
> \n on >
From the table, = 1.28 and = 1.64
/ /
: − 1.28; = 45 (1) and : + 1.64; = 75 (2)
Solving equations (1) and (2), : = 58.15 and ; = 10.28
aM > J > on > aM n^."n on n^."n
æ,hZ ,[ u ,[ &ki[, l_([[ = 60 < < 75 = 4 < < 9= 4 <á< 9
/ / / "M.%^ "M.%^
= 0.18 < á < 1.64 = 0 < á < 1.64 − 0 < á < 0.18
= 0.4495 − 0.0714 = 0.3781
Percentage of students getting first class = 38
Now percentage of students getting second class = 100 – (students who have failed, got 1st class and got distinction)
Percentage of students getting second class = 100 − 10 + 38 + 5 = 47.
11
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
5. If the actual amount of instant coffee which a filling machine puts into ‘6 – ounce’ jars is a RV having a normal
distribution with S.D. is 0.05 ounce and if only 3% of the jars are to contain less than 6 ounce of coffee, what
must be the mean fill of these jars?
Solution: Let X be the actual amount of coffee put into the jars. Then X follows N μ, σ , σ = 0.05
] a > > a
< 6 = 3% = = 0.03 ⇒ 4−∞ < á < M.Mn9 = 0.03 ⇒ 40 < á < M.Mn9 = 0.5 − 0.03 = 0.47
"MM
> a
From the table, = 1.808, Ý = “. G•G’ —µC•O¸
M.Mn
6. In a newly constructed township, 2000 electric lamps are installed with an average life of 1000 burning hours and
standard deviation of 200hours. Assuming the life of the lamps follows normal distribution, find (i)The no. of
lamps expected to fail during the first 700 hrs. (ii) In what period of burning hours 10% of the lamps fail.
Solution: : = 1000, ; = 200
J > oMM "MMM
(i) ≤ 700 = 4 / < %MM 9 = á < −1.5 = á > 1.5 = 0.5 − 0 < á < 1.5
= 0.5 − 0.4332 = 0.0668 (∵From Normal Table)
The no. of lamps that fail to burn in the first 700 hours = 2000 × 0.0668 = 133.6 ≈ 134
(ii) Let , be the period at which 10% of lamps fail.
J > "MMM "MMM
≤ , = 0.1 ⇒ 4 ≤ 9 = 0.1 ⇒ 4á ≥ 9 = 0.1
/ %MM %MM
"MMM "MMM
40 ≤ á ≤ %MM 9 = 0.5 − 0.1 = 0.4 ⇒ %MM
= 1.28 ⇒ = ë’’ (∵From Normal Table)
7. The marks obtained by a number of students in a certain subject are approximately normally distributed with
mean 65 and standard deviation 5. If 3 students are selected at random from this group, what is the probability
that at least 1 of them would have scored above 75?
Solution: : = 65, ; = 5
J > on > on an
> 75 = 4 > 9 = 4á > n 9 = á > 2 = 0.5 − 0 < á < 2 = 0.5 − 0.4772 = 0.0228
/ /
= ( [,hZ , [lgi [ ('gj 75 = 0.0228 , = 1 − = 0.9772 and = 3
t = Ö Ö Ö
, t = 0,1, … , t = 3 Ö 0.0228 Ö 0.9772 ] Ö , t = 0,1, …
Ô ≥ 1 = 1 − Ô < 1 = 1 − 0 = 1 − 3 M 0.0228 M 0.9772 ] M = 0.0667
8. In a normal distribution 31% of the items are under 45 and 8% are over 64. Find the mean and S.D.
Solution: Let mean by μ and standard deviation σ
31% of the items are under 45 and 8% are over 64
< 45 = 31% and > 64 = 8%
J > \n > J > a\ >
4 < 9 = 0.31 and 4 > 9 = 0.08
/ / / /
\n > a\ >
4á < /
9 = 0.31 and 4á > / 9 = 0.08
> \n a\ >
40 < á < /
9 = 0.5 − 0.31 = 0.19 and 40 <á<
/
9 = 0.5 − 0.08 = 0.42
> \n a\ >
/
= 0.5 and /
= 1.4
: − 0.5; = 45 and : + 1.4; = 64
Solving for μ and σ, we get ì = œG and í = HG
9. In a distribution exactly normal 7% of the items are under 35 and 89% are under 63. What are the mean and S.D.
of the distribution?
Solution: Let mean by μ and standard deviation σ
7% of the items are under 35 and 89% are over 63
< 35 = 7% and < 63 = 89%
J > ]n > J > a] >
4 < 9 = 0.07 and 4 < 9 = 0.89
/ / / /
]n > a] >
4á < /
9 = 0.07 and 4á < / 9 = 0.89
> ]n a] >
40 < á < /
9 = 0.5 − 0.07 = 0.43 and 40 <á<
/
9 = 0.89 − 0.5 = 0.39
> ]n a] >
/
= 1.48 and /
= 1.23
: − 1.48; = 35 and : + 1.23; = 63
Solving for μ and σ, we get ì = œG. Ž and í = HG. ŽŽ
12
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
10. In a normal distribution of a large group of men 5% are under 60 in height and 40% are between 60 and 65. Find
the mean height and S.D.
Solution: Let mean by μ and standard deviation σ
5% of the items are under 60 and 40% are between 60 and 65
< 60 = 5% and 60 < < 65 = 40% = 0.4
J > aM >
4 < 9 = 0.05 and < 65 = < 60 + 60 < < 65
/ /
aM > J > an >
4á < /
9 = 0.05 and 4 / < / 9 = 0.05 + 0.4 = 0.45
> aM > an
40 < á < /
9 = 0.5 − 0.05 = 0.45 and 40 < á <
/
9 = 0.5 − 0.45 = 0.05
> aM > an
= 1.645 and = 0.13
/ /
: − 1.645; = 60 and : + 0.13; = 65
Solving for μ and σ, we get ì = “œ. ’V and í = Ž. V•
11. If X is î Ž, ’ . Find Á so that A | − Ž| > ¾ = G. Gœ.
J > J ]
Solution: Let mean by μ and standard deviation σ. : = 3, ; % = 4, ; = 2, á = /
= %
J ] ¿ ¿ ¿
| − 3| > ¾ = 0.05 ⇒ 4ï ï > %9 = 0.05 ⇒ 4|á| > %9 = 0.05 ⇒ 2 4á > %9 = 0.05
%
¿ ¿ ¿ ¿
4á > % 9 = 0.025 ⇒ 0.5 − 40 < á < % 9 = 0.025 ⇒ 40 < á < %9 = 0.475 ⇒ % = 1.96 ⇒ Á = Ž. •V
12. The percentage of a particular compound contained in a rocket fuel follows the distribution î ŽŽ, Ž , through
the specification for is that it should lie between 30 and 35. The manufacturer will get a net profit (per unit of
the fuel) of Rs. 100, if ŽG < < 35, Rs. 50, if Vœ < ≤ ŽG —• Žœ ≤ < 40 and incur a loss of Rs. 60 per unit
of the fuel otherwise. Find the expected profit of the manufacturer. If he wants to increase his expected profit by
50% by increasing the net profit on that category of the fuel that meets the specification, what should be the new
net profit per unit of the fuel of this category?
Solution: r :, ; , r 33, 3 , : = 33, ; = 3
]M > J > ]n > ]M ]] ]n ]]
30 < < 35 = 4 < < 9 = 4 <á< 9 = −1 < á < 0.67
/ / / ] %
= −1 < á < 0 + 0 < á < 0.67 = 0<á<1 + 0 < á < 0.67
= 0.3413 + 0.2486 = 0.5899 (∵From Normal Table)
%n > J > ]M > %n ]] ]M ]]
25 < ≤ 30 = 4 < < 9= 4 <á< 9= −2.67 < á < −1
/ / / ] %
= 1 < á < 2.67 = 0 < á < 2.67 − 0 < á < 1 = 0.4962 − 0.3413 = 0.1549
]n > J > \M > ]n ]] \M ]]
35 ≤ < 40 = 4 < < 9= 4 <á< 9= 0.67 < á < 2.33
/ / / ] %
= 0 < á < 2.33 − 0 < á < 0.67 = 0.4901 − 0. .2486 = 0.2415
T 25 < ≤ 30 gi 35 ≤ < 40 U = 25 < ≤ 30 + 35 ≤ < 40 = 0.1549 + 0.2415 = 0.3964
< 25 gi > 40 = 1 − 0.5899 + 0.3964 = 0.0137
Profit / Unit Probability
Rs. 100 0.5899
Rs. 50 0.3964
Rs. -60 0.0137
P ig&k, i h k, = ð[. 100 × 0.5899 + 50 × 0.3964 − 60 × 0.0137 = ð[. 79
Let the revised net profit per unit of the first category fuel be ¾.
P ð jk[ Z ig&k, i h k, = ð[. ¾ × 0.5899 + 50 × 0.3964 − 60 × 0.0137 = ð[. 0.5899¾ + 18.998
P ð jk[ Z ig&k, i h k, = ð[. 79 + ð[. 39.5
""^.n "^.‘‘^
0.5899¾ + 18.998 = 118.5 ⇒ ¾ = = 168.68 ⇒ Á = H“•
M.n^‘‘
All the Best
Dr. P. GodhandaRaman, Assistant Professor, Department of Mathematics
SRM IST, Kattankulathur – 603 203
Email : godhanda@gmail.com, Mobile : 9941740168
13
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Normal Distribution Table
Dr. P. GodhandaRaman
Assistant Professor, Department of Mathematics, SRM University, Kattankulathur
Email : godhanda@gmail.com, Mobile : 9941740168
14
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
UNIT – 3 : TESTING OF HYPOTHESIS
Syllabus:
• Z – Test
• t – Test
• F - Test
• Chi – Square Test – Goodness of Fit, Independent Attributes
Samples
15
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Critical Region : A region, corresponding to a statics ,, in the sample space S which amounts to rejection of the null
hypothesis óM is called as critical region or region of rejection. The region of the sample space S which amounts to
the acceptance óM is called acceptance region.
Level of Significance : The probability ù that a random value of the statistic , belongs to the critical region is known
as the level of significance. In other words, level of significance is the size of the Type I error. The levels of
significance usually employed in testing of hypothesis are 5% and 1%.
LARGE SAMPLES
If the size of the sample ≥ 30, then that sample is called large sample. If is large, the distributions, such as
Binomial, Poisson, chi – square etc., are closely approximated by normal distributions. There are 4 important test to
test the significance of large samples. 1.Single Proportion 2.Difference of proportions 3.Single mean 4. Difference of
means.
PROCEDURE FOR TESTING OF HYPOTHESIS
1. Set up the null hypothesis óM.
2. Set up the alternative hypothesis ó" .
This will enable us to decide whether we have to use a single tailed (right or left) test or two tailed test.
3. Choose the appropriate level of significance(either 5% or 1% level). This is to be decided before sample is drawn.
ú
4. Compute the test statistics Þ = under the null hypothesis.
¼ú
5. We compare the computed value of Þ in step (4) with the tabulated value Þû at given level of significance ù.
If the calculated value of á is less than tabulated value Þû then óM accepted.
If the calculated value of á is greater than tabulated value Þû then óM rejected.
Critical Value àü of à :
Level of Significance ü
Critical Value àü
H% G. GH œ% G. Gœ
Two Tailed Test |áû | = 2.58 |áû | = 1.96
Right Tailed Test áû = 2.33 áû = 1.645
One Tailed Test áû = −2.33 áû = −1.645
SINGLE PROPORTION
Suppose a large sample of size is taken from a normal population. To test the significant difference between the
E A
sample proportion and the population proportion , we use the statistic à =
Aþ
ý
C
º
1. The probable limits for the observed proportion of successes are given by ± Þû ý .
‹Œ
2. If P is not known, the limits for the population proportion P are given by ± Þû ý where =1− .
3. If ù is not given, we can take safely 3; limits. Hence, confidence limits for observed proportion are ±
º ‹Œ
3ý and confidence limits for the population proportion P are ± 3ý where =1− .
‹Œ
4. 95% confidence limits for population proportion P are given by ± 1.96ý where =1−
‹Œ
5. 99% confidence limits for population proportion P are given by ± 2.58ý where =1−
PROBLEMS IN SINGLE PROPORTION
Two Tailed Test
1. A coin is tossed 256 times and 132 heads are obtained. Would you conclude that the coin is a biased one?
J "]%
Solution: = 256, = No. of success = 132, = proportion of successes in the sample = = = 0.5156 ,
%na
"
= populatin proportion = %
, =1− = 0.5 ; Null Hypothesis óM ∶ The coin is unbiased.
Alternative Hypothesis ó" : The coin is biased. ≠ 0.5 (two tailed test)
‹ º M.n"na M.n
Test statistics Þ = = = 0.4992 , |Þ| = 0.4992 < 1.96 ; óM is accepted. Hence the coin is unbiased.
§.¯ §.¯
ý ý
3¯°
16
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
2. In a city, a sample of 1000 people were taken & out of them 540 are vegetarians & the rest are non vegetarians.
Can we say that both habits of eating are equally popular in the city at 1% & 5% level of significance?
n\M
Solution: = 1000, = sample proportion of vegetarians = = 0.54
"MMM
"
= populatin proportion of vegetarians = = 0.5, %
=1− = 0.5
óM ∶ = 0.5 Both habits are equally popular in the city). ó" : ≠ 0.5 (two tailed test)
‹ º M.n\ M.n
Test statistics Þ = = = 2.5298
§.¯ §.¯
ý ý
2§§§
(i) |Þ| = 2.5298 < 2.58, óM accepted at 1%. (ii) |Þ| = 2.5298 > 1.96, óM rejected at 5% level of significance.
Both types of eaters are popular at 1% level and not so at 5% level of significance.
Right Tailed Test
3. In a big city 325 men out of 600 men were found to be smokers. Does this information support the conclusion that
the majority of men in this city are smokers?
]%n
Solution: = 600, = aMM = 0.5417, = Population proportion of smokers in the city = 0.5, = 0.15
Null Hypothesis óM ∶ = 0.5, Alternative Hypothesis ó" : > 0.5 (right tailed test)
‹ º M.n\"o M.n
Test statistics Þ = = = 2.04 , Þû = 1.645 at 5% level of significance
§.¯ §.¯
ý ý
°§§
5. The fatality rate of typhoid patients is believed to be 17.26 per cent. In a certain year 640 patients suffering from
typhoid were treated in a metropolitan hospital and only 63 patients died. Can you consider the hospital efficient?
a]
Solution: = 640, = a\M = 0.0984, = 0.1726, =1− = 0.8274
Null Hypothesis óM ∶ = , i.e. The hospital is not efficient.
Alternative Hypothesis ó" : < . (Left tailed test)
‹ º M.M‘^\ M."o%a
Test statistics Þ = = = −4.96 , |Þ| = 4.96 > −2.33 , óM is rejected at 1% level of significance.
§.2±3° §.¬3±®
ý ý
°®§
The hospital is efficient in bringing down the fatality rate of typhoid patients.
Confidence Limits for the Population Proportion
6. A random sample of 500 toys was taken from a large consignment and 65 were found to be defective. Find the
percentage of defective toys in the consignment.
an
Solution: = 500, = nMM = 0.13, =1− = 0.87, Confidence Limits for the Population Proportion P:
‹Œ M."]×M.^o
= ± 3ý = 0.13 ± 3ý = 0.175 ( Z 0.085
nMM
The percentage of defective toys in the consignment lies between 17.5 and 8.5.
7. A biased coin was thrown 400 times and 240 heads turned up. Find the probability of throwing heads in a single
trial almost certainly lies between 0.53 and 0.67.
%\M
Solution: = 400, = \MM = 0.6, =1− = 0.4 Confidence Limits for the Population Proportion P :
‹Œ M.a×M.\
= ± 3ý = 0.6 ± 3ý = 0.6735 ( Z 0.5265
\MM
The percentage of defective toys in the consignment lies between 52.65 and 67.35.
17
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
DIFFERENCE OF PROPORTIONS
Suppose 2 large samples of sizes "& % are taken respectively from 2 different populations. To test the significant
EH EV CH EH ?CV EV
difference between the sample proportions and find à = where A = and þ = H − A.
" % H H CH ?CV
ýAþ= ? @
CH CV
óM is accepted. i.e. men and women do not differ significantly in their attitude as regards the proposal.
2. In a large city A, 20% of a random sample of 900 school boys had a slight physical defect. In another large city B,
18.5% of a random sample of 1600 school boys had the same defect. Is the difference between the proportions
significant?
Solution: " = 900, % = 1600, " = 20% = 0.2, % = 18.5% = 0.185
Null Hypothesis óM ∶ " = % , Alternative Hypothesis ó" : " ≠ % (two tailed test)
‹2 ‹3 2 ‹2 ? 3 ‹3
Test statistics Þ = , where = , =1−
ýº 4
2
?
2
9 2? 3
2 3
‘MM×M.% ? "aMM×M."^n "^M?%‘a
= = = 0.1904, =1− = 0.8096
‘MM?"aMM %nMM
‹2 ‹3 M.% M."^n
Þ= 2 2
= = 0.92
2 2
ýº 4 ? 9 ý M."‘M\×M.^M‘a 4 ? 9
2 3 ²§§ 2°§§
|Þ| < Þû , óM is accepted. Therefore, the difference between " and % is not significant at 5% level.
Right Tailed Test
4. Out of a sample of 1000 persons were found to be coffee drinkers. Subsequently, the excise duty on coffee was
increased. After the increase in excise duty of coffee seeds, 800 people were found to take coffee out of a sample
1200. Test whether there is any significant decrease in the consumption of coffee after the increase in excise duty
^MM ^MM
Solution: " = 1000, % = 1200, " = = 0.8, % = = 0.67
"MMM "%MM
Null Hypothesis óM ∶ " = %, Alternative Hypothesis ó" : " > % (right tailed test)
‹2 ‹3 2 ‹2 ? 3 ‹3
Test statistics Þ = , where = = 0.7273, =1− = 0.2727
ýº 4
2
?
2
9 2? 3
2 3
18
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
M.^ M.ao
Þ= = 6.82, Þû = 2.33 at 1% level of significance, |Þ| > Þû , óM is rejected.
2 2
ý M.o%o]×M.%o%o 4 ? 9
2§§§ 23§§
That is, there is significant decrease in the consumption of tea after the increase in duty.
5. A machine produced 20 defective units in a sample of 400. After overhauling the machine, it produced 10
defective units in a batch of 300. Has the machine improved in production due to overhauling. Test at 5% level of
significance.
%M "M
Solution: " = 400, % = 300, " = = 0.05, % = = 0.033
\MM ]MM
Null Hypothesis óM ∶ " = %, Alternative Hypothesis ó" : " > % (right tailed test)
‹2 ‹3 2 ‹2 ? 3 ‹3
Test statistics Þ = , where = = 0.0427, =1− = 0.9573
ýº 4
2
?
2
9 2? 3
2 3
M.Mn M.M]]
Þ= = 1.1 , Þû = 1.645 at 5% level of significance, |Þ| < Þû , óM is accepted.
2 2
ý M.M\%o×M.‘no] 4 ? 9
®§§ •§§
Note: If the samples have been drawn from the same population then ;"% = ;%% = ; % , Þ = 2 3
83 83
ý ?
2 3
3 3
2 ¼2 ? 3 ¼3
If ; is not known we can use a estimate of ; given by ; = % %
2? 3
Two Tailed Test
1. In a random sample of size 500, the mean is found to be 20. In another independent sample of size 400, the mean
is 15. Could the samples have been drawn from the same population with S.D. 4?
Solution: " = 500, % = 400, " = 20, % = 15, ; = 4
Null Hypothesis óM ∶ ̅" = ̅% , Alternative Hypothesis ó" : ̅" ≠ ̅% (two tailed test)
̅2 ̅3 %M "n
Test statistics Þ = 2 2
= = 18.6 ; Þû = 2.58 at 1%, |Þ| > Þû , óM is rejected.
2 2
/ ý ? \ ý ?
2 3 ¯§§ ®§§
That is, the samples could not have been drawn from the same population.
2. The mean of 2 large samples 1000 and 2000 members are 67.5 inches and 68.0 inches respectively. Can the
samples the regarded as drawn from the same population of S.D. 2.5 inches.
Solution: " = 1000, % = 2000, " = 67.5, % = 68, ; = 2.5
Null Hypothesis óM ∶ ̅" = ̅% , Alternative Hypothesis ó" : ̅" ≠ ̅% (two tailed test)
̅2 ̅3 ao.n a^
Test statistics Þ = 2 2
= = −5.1, |Þ| = 5.1
2 2
/ ý ? %.n ý ?
2 3 2§§§ 3§§§
|Þ| < Þû , óM is accepted. That is, we cannot conclude that boys perform better than girls.
20
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
SMALL SAMPLES C < 30
• t – test
• F – test
• øV − O¸ DÇš Fµ–•O − O¸
Students ‘t’ – Test - SINGLE MEAN
B Ý ∑ 2
Standard deviation given directly: = ¸ , where ̅ = ,
4 9
√C6H
̅ - sample mean, : – population mean, – sample size, [ % – sample variance
B Ý ∑C
š H Bš B
V
Standard deviation not given directly: = , where V
=
4 9 C H
√C
̅ - sample mean, : – population mean, – sample size, æ % – population variance ; % , Degree of freedom : C − H
DIFFERENCE MEAN
B ∑ 2 ∑ 2Ö
= , where = , t = , Degree of freedom : CH + CV − V
ý V = H ? H @
CH CV
© 2 ¼23 ? 3 ¼33 ª
Standard deviation given directly: [ % =
2? 3 %
∑ ̅ 3? ∑ Ö Ö 3
Standard deviation not given directly: [% = ,
2? 3 %
Confidence or Fiducial Limits for Ý:
95% Confidence limits for μ : ̅ − ,M.Mn 4 9 ≤ : ≤ ̅ + ,M.Mn 4 9
√ √
99% Confidence limits for μ : ̅ − ,M.M" 4 9 ≤ : ≤ ̅ + ,M.M" 4 9
√ √
Application of t – distribution:
The t – distribution has a wide number of applications in statistics, some of which are enumerated below
(i) To test if the sample mean ̅ differs significantly from the hypothetical value μ of the population mean.
(ii) To test the significance of the difference between two sample means.
(iii) To test the significance of an observed sample correlation coefficient and sample regression coefficient
(iv) To test the significance of observed partial correlation coefficient.
Assumption for students’s t – test:
(i) The parent population from which the sample is drawn is normal.
(ii) The sample observations are independent, that is, the sample is random.
(iii) The population standard deviation ; is unknown.
PROBLEMS
Students ‘t’ – Test : Single Mean and Standard Deviation Given Directly
1. The mean weekly sales of soap bars in departmental stores was 146.3 bars per store. After an advertising
campaign the mean weekly sales in 22 stores for a typical week increased to 153.7 and showed a S.D. of 17.2. Was
the advertising campaign successful?
Solution: = 22, ̅ = 153.7, : = 146.3 , [ = 17.2
Null Hypothesis òG ∶ : = 146.3, i.e. The advertising campaign is not successful.
Alternate Hypothesis òH : : > 146.3 (Right tailed)
̅ > "n].] "\a.]
,= Â = 2±.3 = 9.03, degree of freedom: n − 1 = 21 at 5% level of significance = 1.72.
4 9 4 9
√ 62 √3362
Calculate value t > Tabulated t. G is rejected.
Conclusion: The advertising campaign was definitely successful in promoting sales.
2. A sample of 20 items has mean 42 units and S.D. 5 units. Test the hypothesis that it is a random sample from a
normal population with mean 45 units.
Solution: = 20, ̅ = 42, : = 45, [ = 5
Null Hypothesis òG ∶ : = 45 , Alternate Hypothesis òH : : ≠ 45 (Two tailed)
̅ > \% \n
,= Â = ¯ = 2.615, degree of freedom: n − 1 = 19 at 5% level of significance = 2.09.
4 9 4 9
√ 62 √3§62
Calculate value t > Tabulated t. G is rejected. The sample could not have come from this population.
21
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
3. A spare part manufacturer is making spare parts with axle diameter of 0.700 inch. A random sample of 10 parts
shows a mean diameter of 0.742 inch with a S.D. of 0.040 inch. Verify whether the work satisfies the
specifications.
Solution: = 10, ̅ = 0.742, : = 0.700, [ = 0.040
Null Hypothesis òG ∶ : = 0.700 , Alternate Hypothesis òH : : ≠ 0.700 (Two tailed)
̅ > M.o\% M.oMM
,=  = §.§®§ = 3.15, degree of freedom: n − 1 = 9 at 5% level of significance = 2.26.
4 9 4 9
√ 62 √2§62
Calculate value t > Tabulated t. G is rejected. Conclusion: The product is not meeting the specifications.
4. The mean life time of a sample of 25 bulbs is found as 1550 hours with a S.D. of 120 hours. The company
manufacturing the bulbs claims that the average life of their bulbs is 1600 hours. Is the claim acceptable at 5%
level of significance?
Solution: = 25, ̅ = 1550, : = 1600, [ = 120
Null Hypothesis òG ∶ ̅ = : , Alternate Hypothesis òH : ̅ < : (Left Tailed)
̅ > "nnM "aMM
,= Â = 23§ = −2.04, degree of freedom: n − 1 = 24 at 5% level of significance = 1.71.
4 9 4 9
√ 62 √3¯62
Calculate value t > Tabulated t. G is rejected. The claim of the company cannot be accepted at 5% LOS.
Students ‘t’ – Test : Single Mean and Standard Deviation Not Given Directly
5. A random sample of 10 boys had the following I.Q.’s 70, 120, 110, 101, 88, 83, 95, 98, 107, 100. Do these date
support the assumption of a population mean I.Q.’s of 100? Find a reasonable range in which most of the mean
I.Q. values of samples of 10 boys lie.
Solution: = 10, : = 100, òG : : = 100, òH : : ≠ 100 (Two tail)
∑ 2 ‘o% ∑ 2 ̅ 3 "^]].a
̅= = = 97.2, æ% = = = 203.73
"M " "M "
̅ > ‘o.% "MM %.^
,= = = = −0.62 ⇒ | | = G. “V , d.f.= n − 1 = 10 − 1 = 9 at 5% LOS = 2.262.
4 9 3§•.±• √%M.]o]#
√ ý
2§
B B−B B−B V
70 -27.2 739.84
120 22.8 519.84
110 12.8 163.84
101 3.8 14.44
88 -9.2 84.64
83 -14.2 201.64
95 -2.2 4.84
98 0.8 0.64
107 9.8 96.04
100 2.8 7.84
∑ B = 972 ∑ B − B V =1833.60
Calculate value t < Tabulated t. G is accepted.
Conclusion: The data are consistent with the assumption for a mean I.Q. of 100 in population.
95% Confidence Limits for Ý: ̅ − ,M.Mn 4 9 ≤ : ≤ ̅ + ,M.Mn 4 9
√ √
%M].o] %M].o]
97.2 − 2.262 ý ≤ : ≤ 97.2 + 2.262 ý
"M "M
86.99 ≤ : ≤ 107.41, Hence the required 95% confidence interval is T86.99, 107.41U.
6. The wages of 10 workers taken at random from a factory are given as Wages: 578, 572, 570, 568, 572, 578, 570,
572, 596, 584. Is it possible that the mean wage of all workers of this factory could be Rs. 580
Solution: = 10, : = 580, òG : : = 580, òH : : ≠ 580 (Two tailed)
∑ 2 noaM ∑ 2 ̅ 3 ana
̅= = = 576, æ% = = = 72.89
"M " "M "
̅ > noa n^M \
,= = = = −1.48 ⇒ | | = H. ’ ;
4 9 ±3.¬² √o.%^‘#
√ ý
2§
23
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Null Hypothesis óM : BH = BV , i.e. There is no significant difference between the mean increase in weight due to
diets A & B, Alternate Hypothesis : ó" : BH ≠ BV (Two tailed)
∑ 2 ]]a ∑ 2Ö \nM
" = 12, % = 15, = = "%
= 28, t = = "n
= 30
∑ ̅ 3? ∑ Ö Ö 3
Standard deviation not given directly: [ % = = 71.6
2? 3 %
̅ Ö %^ ]M
,= = = −0.609 ⇒ | | = G. “G•, d.f.= " + % − 2 = 25 at 5% LOS = 2.06.
ý 3 4 2 ? 2 9 2 2
ýo".a 4 ? 9
2 3 23 2¯
24
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
3 "" M.^ 3 3 ‘ M.n 3 3
2 ¼2 3 ¼3 M.oM\
æ"% = = = 0.704 , æ%% = = = 0.28125, Õ = 2
3 = = 2.503,
2 " "" " 3 " ‘ " 3 M.%^"%n
d.f. = " − 1, % − 1 = 10, 8 at 5% LOS = 3.34. Calculate value F < Tabulated F. G is accepted.
14. The time taken by workers in performing a job by Method I and Method II is given below
Method I 20 16 26 27 23 22 -
Method II 27 33 42 35 32 34 38
Do the data show that the variances of time distribution form population from which these samples are drawn do
not differ significantly?
∑ 2 "]\ ∑ 2Ö %\"
Solution: " = 6, % = 7, = = = 22.3, t = = = 34.4
2 a 3 o
B B−B B−B V − − V
20 -2.3 5.29 27 -7.4 54.76
16 -6.3 39.69 33 -1.4 1.96
26 3.7 13.69 42 7.6 57.76
27 4.7 22.09 35 0.6 0.36
23 0.7 0.49 32 -2.4 5.76
22 -0.3 0.09 34 -0.4 0.16
∑ B = HŽ’ ∑ B−B V
= H. Ž’
38 3.6 12.96
∑ = V’H ∑ − V = HŽV. ëV
Null Hypothesis : óM : ;J% = ;Ó% , Alternate Hypothesis : ó" : ;J% ≠ ;Ó%
∑ ̅ 3 ^".]\ ∑ Ö Ö 3 "]].o% 3 %%.%‘
æJ% = = = 16.268, æÓ% = = = 22.29 , Õ = 3 = "a.%a^ = 1.37,
2 " a " 3 " o "
d.f. = % − 1, " − 1 = 6, 5 at 5% LOS = 4.95. Calculate value F < Tabulated F. G is accepted.
15. Two horses A and B were tested according to the time (in seconds) to run a particular track with the following
Horse A 28 30 32 33 33 29 34
Horse B 29 30 30 24 27 29 -
Test whether the two horses have the same running capacity.
∑ 2 %"‘ ∑ 2Ö "a‘
Solution: " = 7, % = 6, = = = 31.28, t = = = 28.2
2 o 3 a
B B−B B−B V − − V
25
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
∑ 2 "%] ∑ 2Ö "o\
Solution: = 5, % = 6, = = = 24.6, t = = = 29
"
2 n 3 a
B B−B B−B V − − V
24 -0.6 0.36 27 -2 4
27 2.4 5.76 30 1 1
26 1.4 1.96 28 -1 1
21 -3.6 12.96 31 2 4
25 0.4 0.16 22 -7 49
∑ B = HVŽ ∑ B − B V = VH. V
36 7 49
∑ = Hë’ ∑ − V
= HG
Student’s t test
Null Hypothesis : óM: BH = BV , Alternate Hypothesis : ó" : BH ≠ BV (Two tailed)
∑ ̅ 3? ∑ Ö Ö 3 %".%?"M^ ̅ Ö %\.a %‘
[% = = = 14.35, , = = = −1.92 ⇒ | | = H. •V,
2? 3 % n?a %
ý 3 4 2 ? 2 9 ý"\.]n 4 ? 9
2 2
2 3 ¯ °
26
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
^]
Total number of accidents = 83. Based on óM , the expected number of accidents on any day = = 14.
a
Degree of freedom = − 1 = 6 − 1 = 5 at 5% LOS = 11.07
% %−
%
%−
V
V
14 14 0 0 0
18 14 4 16 1.143
12 14 -2 4 0.286
11 14 -3 9 0.643
15 14 1 1 0.071
14 14 0 0 0
2.143
%š
= V. H’Ž, Calculated $ % <tabulated $ % , òG is accepted. The accidents are uniformly distributed.
V
øV = ∑CšLH š
š
17. The theory predicts the proportion of beans, in the four groups A, B, C, and D should be 9 : 3 : 3 : 1. In an
experiment with 1600 beans the numbers in the four groups were 882, 313, 287, and 118. Does the experimental
Solution: HM : The experimental result support the theory. d.f. = − 1 = 4 − 1 = 3 at 5% LOS = 7.81
result support the theory.
% %− %− V % V
9
× 1600 = 900
16
882 -18 324 0.36
3
× 1600 = 300
16
313 13 169 0.563
3
× 1600 = 300
16
287 -13 169 0.563
1
× 1600 = 100
16
118 18 324 3.24
4.726
27
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
90 × 100 90 × 100
P 60 = = 45 P 30 = = 45
200 200
90
% %−
%
%−
V
V
60 45 15 225 5
30 45 -15 225 5
40 55 -15 225 4.09
70 55 15 225 4.09
18.18
%š V
øV = ∑CšLH š
= H . H , d.f. = i − 1 l − 1 = 2 − 1 2 − 1 = 1 at 5% LOS = 3.841.
š
Calculated $ > tabulated $ % , òG is rejected. New and conventional treatment are not independent.
%
19. Given the following contingency table for hair colour and eye colour. Find the value of chi square. Is there good
association between the two.
Hair Colour
Total
Fair Brown Black
Blue 15 5 20 40
Eye Colour Grey 20 10 20 50
Brown 25 15 20 60
Total 60 30 60 150
Solution: HM : The two attributes Hair colour and eye colour are independent.
% %− %− V % V
15 16 -1 1 0.0625
5 8 -3 9 1.125
20 16 4 16 1
20 20 0 0 0
10 10 0 0 0
20 20 0 0 0
25 24 1 1 0.042
15 12 3 9 0.75
20 24 -4 16 0.666
3.6458
60 × 40 30 × 40 60 × 40
P 15 = = 16 P 5 = =8 P 20 = = 16
150 150 150
40
60 × 50 30 × 50 60 × 50
P 20 = = 20 P 10 = = 10 P 20 = = 20
150 150 150
50
60 × 60 30 × 60 60 × 60
P 25 = = 24 P 15 = = 12 P 20 = = 24
150 150 150
60
60 30 60 150
%š V
øV = ∑CšLH š
= Ž. “’œ , d.f. = i − 1 l − 1 = 3 − 1 3 − 1 = 4 at 5% LOS = 9.488.
š
Calculated $ % < tabulated $ % , òG is accepted. The hair colour and eye colour are independent.
All the Best
Dr. P. Godhanda Raman, Assistant Professor(Senior Grade), Department of Mathematics, SRM IST, Kattankulathur
Email : godhanda@gmail.com, Mobile : 9941740168
28
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
t – test table
29
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
F test Table
30
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com
Chi-Square Test Table
Regards!
Dr. P. Godhanda Raman, Assistant Professor(Senior Grade), Department of Mathematics, SRM IST, Kattankulathur
Email : godhanda@gmail.com, Mobile : 9941740168
.
31
Dr. P. Godhandaraman, Assistant Professor, Dept of Maths, SRM IST, Kattankulathur, Mobile: 9941740168, Email: godhanda@gmail.com