Business Research and Analytics (CORP 2181)
Lecture 4: Normal Distribution
Dr . Rahul Jain
Leicester Castle Business School
Learning Objectives
- To understand the topic on Normal Distribution and its
importance
- Draw normal distribution curves and calculate the standard
score (z score)
• Apply the basic knowledge of normal distribution to solve
problems.
• Interpret the results of the problems.
Recap
Probability = the likelihood of an occurrence
The probability of getting one head upon flipping a coin: 50% head & 50% tail.
• Numerical/ Quantitative data:
1. Discrete (Coin tosses/ Heads or Tails)
2. Continuous (How about continuous variables? How to calculate probability of
specific weight value???) Can we list all the values for weight in a sample? How
many decimals should we take into account?
Probability Density Function
Weight of students in a school
The area under the curve shows the
Probability that a student weight is
between a range of possible values
(25kg – 35 kg). = p-value 25 kg 35 kg
Normal distribution
• Known as Gaussian distribution
• Normal distribution is the most important probability distribution in
statistics.
• It fits many real and natural phenomena.
• E.g. Weight of cereal boxes, height of students, exam scores, blood
pressure and etc..
Difference between parameters and statistics
• Normal distribution:
• Data is obtained from sample or population
• Parameter:
• A number that describes the data from a population
Population mean Population standard deviation
• Statistics:
• A number that describes the data from a sample
Sample mean Sample standard deviation
~
Normal Distribution
1. The normal distribution is usually seen as Bell
shape.
2. It is not skewed in any way, and has that famous
bell shape.
3. The area under the normal distribution is
symmetrical around the mean. Half of the data will 99.8%
be less than the mean and half of them will be. 95.44%
4. The mean, median, and mode are all equal.
5. Most of the data lies near the mean (one single 68.26%
peak).
6. The percentage of data falling on either side of the
mean is consistent.
7. The whole area under the curve equals 1.
σ σ σ
~
σ σ σ
μ
Normal Distribution 99.8%
1. 68.26% of the data under the bell curve occur 95.44
%
within 1 standard deviation (1σ) from the mean. 68.26
%
2. 95.44 of the data under the bell curve occur within
2 standard deviation (2σ) from the mean.
3. 99.80 of the data under the bell curve occur within
3 standard deviation (3σ) from the mean.
σ σ σ σ σ σ
Note that all data are no modeled by the normal
μ
distribution and the distribution is not symmetric.
A right-skewed A left-skewed
distribution has a right distribution has a
tail and the mean is to left tail and the
the right of the peak. mean is to the left of
the peak.
Example
95.44% of the data under the bell curve occur within 2 95.44%
standard deviation (2σ) from the mean.
Assume we want to find the probability of a cereal box
containing greater than 256 gr cereal.
We know that the mean of the weight of the cereal
box is 250gr and the standard deviation is 20 gr.
Mean -2 S.D. = 250-40 = 210 2σ 2σ
210 μ =250 290
95.44% of the weight of the cereal boxes lies
between (210, 290)
Normal Distribution
Assume we want to find the probability of a cereal box
containing greater than 256 gr cereal.
We know that the mean of the weight of the cereal
box is 250gr.
To find the probability of specific value , P-value, we
refer to standard normal distribution tables.
The standard normal distribution tables show the area
under curve for standardized values. So, first we have
to transform the normal distribution to standard
normal distribution.
μ
256
250
Standard normal distribution tables
Standard Normal Distribution
Standard normal distribution = Z distribution
We can transform any normal distribution to
standard normal distribution(Z) using Z score:
μ Z
0
Z ~ N(0,1)
Example – Standardisation
• Assume one production unit produces average of 200 washing
machine per day and the standard deviation is 10 units. X is 220
σ σ σ σ σ σ
X
170 180 190 200 210 220 230
How to obtain the probability of being below/ above
a particular value or between a particular range?
One production unit produces average of 200 washing machine per day and the
standard deviation is 30 units. What is the probability of total production during
one specific day is less than 260 units?
X= variable: total production per day
X = 0.9772
200 260
Z~N(0, )
Z =?
Example
• The average of waiting time in a queue in a fast food restaurant is 5
minutes and the standard deviation is 2 minutes. Variance is 4. Find
the probability of a customer should wait more than 7 minutes and 2
seconds.
X = variable: waiting time X~N(μ, ) X~N(5,4 )
X
5 7.2
= 1- =
Z~N(0, )
Z =?
Questions?
