NORMAL DISTRIBUTION

IS AN APPROXIMATION TO BINIOMIAL
DISTRIBUTION WHEN

- NUMBER OF TRIALS n IS VERY LARGE

- p AND q ARE NOT VERY SMALL

IS DEFINED FOR CONTINUOUS TYPE OF

VARIABLES

NOTE: BINOMIAL DISTRIBUTION TENDS TO THE

FORM OF THE CONTINUOUS CURVE WHEN n
BECOMES LARGE.
 IS A LIMITING CASE OF POISSON DISTRIBUTION
WHEN IT’S MEAN IS LARGE

 IT IS AN EXTENSION OF BOTH BINOMIAL AND

POISSON DISTRIBUTION WHEN n IS LARGE
GRAPH OF NORMAL DISTRIBUTION /
PROPERTIES OF NORMAL CURVE

1) BELL SHAPED SYMMETRICAL CURVE, SYMMETRY

ABOUT MEAN.

2) TAILS NEVER TOUCH X AXIS AND EXTENDS TO

INFINITY IN BOTH DIRECTIONS

3) THE CURVE IS UNIQUE FOR A PARTICULAR PAIR

OF MEAN AND STANDARD DEVIATION.
4) MEAN, MEDIAN, MODE ARE ALL EQUAL AND
COINCIDE AT THE CENTR OF THE CURVE ABOUT
WHICH THE CURVE IS SYMMETRIC.

5) AREA UNDER THE NORMAL CURVE IS

DISTRIBUTED AS FOLLOWS
a) MEAN ± 1σ COVERS 68.27 % AREA
b) MEAN ± 2σ COVERS 95.45 % AREA
c) MEAN ± 3σ COVERS 99.73 % AREA
 THE NORMAL DISTRIBUTION CURVE

68.27%
f( )
z

95.45%

-3σ -2σ -σ MEAN σ 2σ 3σ

99.73%
 PROBABILITIES AND AREAS UNDER
NORMAL CURVE

• THE PROBABILITY THAT A NORMAL VARIABLE

WILL BE WITHIN 1 STANDARD DEVIATION FROM
ITS MEAN (ON EITHER SIDE) IS 0.6827
• THE PROBABILITY THAT A NORMAL VARIABLE
WILL BE WITHIN 2 STANDARD DEVIATIONS FROM
ITS MEAN IS 0.9545
• THE PROBABILITY THAT A NORMAL VARIABLE
WILL BE WITHIN 3 STANDARD DEVIATION FROM
ITS MEAN IS 0.9973
STANDARD NORMAL PROBABILITY
DISTRIBUTION
 PORTIONS OF THE AREA UNDER THE NORMAL
CURVE THAT ARE CONTAINED WITHIN ANY
NUMBER OF STANDARD DEVIATIONS FROM THE
MEAN CAN BE EXTRACTED FROM STATISTICAL
TABLES

 FOR EACH PAIR OF MEAN AND S.D THERE EXISTS

AN UNIQUE CURVE AND THUS A NEED OF UNIQUE
STATISTICAL TABLE (NOT POSSIBLE)

 CONVERT ANY GIVEN NORMAL DISTRIBUTION TO

STANDARD NORMAL DISTRIBUTION(MEAN=0; S.D =1
 CONVERSION PROCESS…..

TRANSFORMATION RULE:

X −X
Z=
σ

 VARIABLE Z MEASURES DEVIATION FROM MEAN

IN UNITS OF STANDARD DEVIATION.

 Z IS CALLED STANDARDISED VARIABLE AND ITS

VALUE IS CALLED STANDARD SCORE.
 STANDARD SCORE GIVES US THE NUMBER OF
STANDARD DEVIATIONS A PARTICULAR LIES
BELOW OR ABOVE THE MEAN

Z=1.8 IMPLIES X= X +1.8 σ

MEANING
X IS 1.8 σ TO THE RIGHT OF MEAN IN X- SCALE
NORMAL CURVE
MEAN = X
68.27% S.D = σ

95.45%
X
-3σ -2σ -σ σ 2σ 3σ
99.73%

STANDARD
NORMAL MEAN=0
CURVE 68.27% S.D=1
95.45%
Z=0
-3 -2 -1 1 2 3
99.73%
 X

GIVEN MEAN=82, S.D =1.5

P[80 .8 < X < 83 .2]

= P[(80 .8 − 82 ) / 1.5 < Z < (83 .2 − 82 ) / 1.5]
= P[−.8 < Z < .8] = .7881 − .2119 = .5762
 TABLE OF Z- TRANSFORM
 GIVES AREA UNDER CURVE FOR +VE SIDE ONLY.
[BETWEEN 0 TO Z].

 FOR AREA ON –VE SIDE, FIND AREA ON +VE SIDE

FOR THE SAME VALUE (SYMMETRIC CURVE)
P(0 ≤ Z ≤ 1.20)= P(-1.20 ≤ Z ≤ 0)

0 z
FIND AREA UNDER THE NORMAL CURVE IN
EACH OF THE FOLLOWING CASES

1) P( 0 ≤ Z ≤ 1.2)

ANS: .3849

0 1.2
2) P( -.68 ≤ Z ≤ 0)

= P( 0 ≤ Z ≤ .68)= .2518

3) P( -.46 ≤ Z ≤ 2.21)

-.46 2.21
P( -.46 ≤ Z ≤ 0) +P( 0≤ Z ≤ 2.21)
= .1772 + .4864 = .6636
4) P( .81 ≤ Z ≤ 1.94 )

.81 1.94

P( 0 ≤ Z ≤ 1.94) – P( 0 ≤ Z ≤ .81)
=.4738 - .2910 = .1828
5) TO THE LEFT OF Z = - .6

=.5 - P(0 ≤ Z ≤ .6)

ANS: .2743

-.6
6) TO THE RIGHT OF Z= -1.28

=.5 + P(0 ≤ Z ≤ 1.28)

ANS: .8997

-1.28
7) RIGHT OF Z= 2.05 AND LEFT OF Z= -1.44

-1.44 2.05

=1- P(-1.44 ≤ Z ≤ 2.05)

= 1-[P(0 ≤ Z ≤ 1.44)+ P(0 ≤ Z ≤ 2.05)]= .0951
Q. A NORMAL CURVE HAS MEAN AS 20 AND S.D AS 10
FIND AREA BETWEEN 15 AND 40.
HINT: P(15≤ X ≤ 40)
SOL: .6687

Q. ASSUME MEAN HEIGHT OF SOLDIERS TO BE 68.22

INCHES WITH VARIANCE OF 10.8 INCHES. HOW
MANY SOLDIERS IN A REGIMENT OF 1,000 ARE
EXPECTED TO BE OVER SIX FEET TALL.
HINT: 6FEET = 72 INCHES=X, FIND P(X ≥ 72 )x1000
SOL: 125
Q. IN A TEST GIVEN TO 1,000 STUDENTS, THE
AVERAGE SCORE WAS 42 AND SD = 24. FIND
A) NUMBER OF STUDENTS EXCEEDIND A SCORE OF
50.
B) NUMBER OF STUDENTS BETWEEN 30 AND 54
SOL: A) 1000x .3696 B) 383

Q. SUPPOSE THAT THE ACTUAL AMOUNT OF INSTANT

COFFEE WHICH A FILLING MACHINE PUTS INTO “5
OUNCE” JARS IS A FOLLOWING A NORMAL
DISTRIBUTION WITH S.D= .04 OUNCE. IF ONLY 3%
OF THE JARS ARE TO CONTAIN LESS THAN 5 OUNCES
OF COFFEE,WHAT MUST BE THE MEAN FILL OF
THESE JARS?
SOL: σ =.04 .47
P(X≤ 5)=.03
X −X
Z=
σ

5 X

FIND Z S.T AREA B/W O AND Z IS .47

.47
5 −X
Z = = −1.88
.03 .04
⇒X =1.88 ×.04 +5
=5.0752
Z 0
Q. ASSUME IN A DISTRIBUTION EXACTLY NORMAL,
7% OF THE ITEMS ARE UNDER 35 AND 89% ARE
UNDER 63. WHAT IS THE MEAN AND S.D OF THE
DISTRIBUTION ?
SOL:
7%
Z CORRESPONDING TO
X=35 AND AREA .43 ON 35 X 63
LEFT IS –1.48
89%
39%
Z CORRESPONDING TO
X=63 AND AREA .39 ON
RIGHT IS 1.23
7%
MEAN=50.3, S.D=10.33 Z1 Z=0 Z2
43%
Q. OF A LARGE GROUP OF MEN 5% ARE UNDER 60
INCHES IN HEIGHT AND 40% ARE BETWEEN 60 & 65
INCHES. ASSUMING A NORMAL DISTRIBUTION, FIND
MEAN HEIGHT AND S.D.
ANS: MEAN=65.429; S.D= 3.3
 CASE
IN A MANUFACTURING UNIT STEEL RODS ARE
MANUFACTURED TO BE 3 INCHES IN DIAMETER , BUT
ARE ACCEPATABLE IF THEY ARE WITHIN THE ERROR
OF .01. IT IS OBSERVED THAT 15% ITEMS ARE OVER
DIMENSIONED AND 15 % OF THE ITEMS ARE UNDER
DIMENSIONED. FIND THE MEAN AND STANDARD
DEVIATION OF THE DISTRIBUTION.

HINT: 15% OVER DIMENSIONED P( X≥ 3.01)= .15

15% UNDER DIMENSIONED P(X ≤ 2.99)= .15
SOLUTION: MEAN=3 S.D=.00962
 NORMAL APPROXIMATION
Q. USE NORMAL APPROXIMATION TO THE BINOMIAL
DISTRIBUTION TO FIND THE PROBABILITY OF
WINNING AT MOST 70 OF 100 MATCHES BY A TEAM,
WHEN THE PROBABILITY OF WINNING EACH MATCH
IS .75
HINT: p= .75, n=100, MEAN=np=.75*100=75,
S.D= SQRT(npq) = 4.33

CONVERT DISRETE TO CONTINUOUS- APPLY

CONTINUITY CORRECTION
PROB OF WINNING NOT MORE THAN 70 MATCHES
≅ P(X ≤ 70.5)=P(Z ≤ -1.04)= .15
Q. FIND THE PROBABILITY OF GETTING 5 HEADS AND
10 TAILS IN 15 TOSSES OF A FAIR COIN USING
NORMAL DISTRIBUTION. ALSO COMPARE YOUR
ANSWER WITH BINOMIAL DISTRIBUTION .
SOL: p BE THE PROB OF GETTING HEAD IN SINGLE
TOSS =1/2
n=15 ; np=7.5 ; S.D= SQRT(npq)=1.94

NORMAL APPROX: P( X =5) ≅ P(4.5 ≤ X ≤ 5.5)

= P(-1.55≤ Z ≤ -1.03)= .0909
BINOMIAL DISTRIBUTION:

P(X=5)=
n
c5 p 5 q n −5 = .09164 (APPROX. SAME)
CASE
Dr. Arnold of Lifeline Hospital is in charge of
Research department. A new medicine is launched
in the market and is known to be accompanied by
bad side effects in 25% of the patients suffering
from high blood pressure. Lifeline Hospital wants
to carry out it’s own Research before actually
prescribing the drug to the patients. For this purpose
120 patients with high blood pressure were treated
with this new [Link] drug would be rejected
if probability of more than 32 patients suffering
from side effects exceeds .25. Help Arnold in taking
the decision.
SOLUTION

p =.25 ; n =120 ; MEAN = np =30; S.D= 4.75

AFTER CONTINUITY CORRECTION

P( X > 32) EQUIVALENT TO P(X ≥ 32.5)

= P(Z ≥ .53)=.2019