Mode
The mode is the most frequently occurring value in a set of data.
Median
The median is the middle value in an ordered array of numbers. For an array with an odd number
of terms, the median is the middle number. For an array with an even number of terms, the
median is the average of the two middle numbers.
Mean
The arithmetic mean is the average of a group of numbers and is computed by summing all
numbers and dividing by the number of numbers.
Percentiles
Percentiles are measures of central tendency that divide a group of data into 100 parts. There are
99 percentiles because it takes 99 dividers to separate a group of data into 100 parts. The nth
percentile is the value such that at least n percent of the data are below that value and at most
(100 − 𝑛) percent are above that value.
Quartiles
Quartiles are measures of central tendency that divide a group of data into four subgroups or parts.
The three quartiles are denoted as Q1, Q2, and Q3.
The first quartile, Q1, separates the first, or lowest, one-fourth of the data from the upper
three-fourths and is equal to the 25th percentile.
The second quartile, Q2, separates the second quarter of the data from the third quarter.
Q2 is located at the 50th percentile and equals the median of the data.
The third quartile, Q3, divides the first three-quarters of the data from the last quarter
and is equal to the value of the 75th percentile.
Range
The range is the difference between the largest value of a data set and the smallest value of a set.
Interquartile Range
Another measure of variability is the interquartile range. The interquartile range is the range of
values between the first and third quartile. Essentially, it is the range of the middle 50% of the data
and is determined by computing the value of Q3 - Q1.
Subtracting the mean from each value of data yields the deviation from the mean (𝑥 − 𝜇). For
a given set of data, the sum of all deviations from the arithmetic mean is always zero.
The mean absolute deviation (MAD) is the average of the absolute values of the
deviations around the mean for a set of numbers.
∑ |𝑥 − 𝜇|
𝑀𝐴𝐷 =
𝑁
Variance
The variance is the average of the squared deviations about the arithmetic mean for a set of
numbers. The population variance is denoted by 𝜎 2 .
2
∑(𝑥 − 𝜇)2
𝜎 =
𝑁
The mean standard deviation is used in calculation of variability for the population, while the
variance is for the samples.
Standard Deviation
The standard deviation is the square root of the variance. The population standard deviation is
denoted by 𝜎 .
∑(𝑥 − 𝜇)2
𝑆𝐷 = 𝜎 = √
𝑁
Both the MAD and standard deviation provide measures of spread. They measure the average
deviation of the observations from their mean. If the observations are spread out, they will tend
to be far from the mean, both above and below. Some deviations will be large positive numbers,
and some will be large negative numbers. But the squared deviations will all be positive. So both
MAD and SD will be large when the data are spread out and small when the data are close
together.
z Scores
A z score represents the number of standard deviations a value (𝑥) is above or below the mean of a set of
numbers when the data are normally distributed.
𝑥−𝜇
𝑧=
𝜎
If a z score is negative, the raw value (𝑥) is below the mean. If the z score is positive, the raw value (𝑥) is
above the mean.
For example, for a data set that is normally distributed with a mean of 50 and a standard deviation of 10,
suppose a statistician wants to determine the z score for a value of 70. This value (𝑥 = 70) is 20 units
above the mean, so the z value is
70 − 50
𝑧= = +2.00
10
This z score signifies that the raw score of 70 is two standard deviations above the mean.
DISTRIBUTIONS AND SAMPLING
For example, suppose an experiment is to measure the arrivals of automobiles at a turnpike tollbooth
during a 30-second period. The possible outcomes are: 0 𝑐𝑎𝑟𝑠, 1 𝑐𝑎𝑟, 2 𝑐𝑎𝑟𝑠, . . . , 𝑛 𝑐𝑎𝑟𝑠. These numbers
(0, 1, 2, . . . , 𝑛) are the values of a random variable.
Suppose another experiment is to measure the time between the completion of two tasks in a production
line. The values will range from 0 seconds to n seconds.
The outcomes for random variables and their associated probabilities can be organized into distributions.
The two types of distributions are discrete distributions, constructed from discrete random variables, and
continuous distributions, based on continuous random variables.
Discrete distributions:
1. Binomial distribution
2. Poisson distribution
3. Hypergeometric distribution
Continuous distributions:
1. Uniform distribution
2. Normal distribution
3. Exponential distribution
4. t distribution
5. Chi-square distribution
6. F distribution
Uniform Distribution
The uniform distribution, sometimes referred to as the rectangular distribution, is a relatively simple
continuous distribution in which the same height, or 𝑓(𝑥), is obtained over a range of values. It is a probability
distribution that has constant probability. The following probability density function defines a uniform
distribution.
1
, 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑎 ≤ 𝑥 ≤ 𝑏
𝑓(𝑥) = {𝑏 − 𝑎
0, 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑜𝑡ℎ𝑒𝑟 𝑣𝑎𝑙𝑢𝑒𝑠
In a uniform, or rectangular, distribution, the total area under the curve is equal to the product of the length
and the width of the rectangle and equals 1.
(𝑎 + 𝑏)
𝑀𝑒𝑎𝑛 𝑜𝑓 𝑈𝑛𝑖𝑓𝑜𝑟𝑚 𝐷𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛 = 𝜇 =
2
(𝑏 − 𝑎)
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑈𝑛𝑖𝑓𝑜𝑟𝑚 𝐷𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛 = 𝜎 =
√12
Normal Distribution
The normal distribution exhibits the following characteristics.
It is a continuous distribution.
It is a symmetrical distribution about its mean. Each half of the distribution is a mirror
image of the other half.
It is asymptotic to the horizontal axis. In theory, the normal distribution is asymptotic
to the horizontal axis. That is, it does not touch the x-axis, and it goes forever in each
direction. The reality is that most applications of the normal curve are experiments that
have finite limits of potential outcomes.
It is unimodal. It is unimodal in that values mound up in only one portion of the graph—
the centre of the curve.
It is a family of curves. Every unique value of the mean and every unique value of the
standard deviation result in a different normal curve.
Area under the curve is 1. The area under the curve yields the probabilities, so the total
of all probabilities for a normal distribution is 1. Because the distribution is symmetric,
the area of the distribution on each side of the mean is 0.5.
A mechanism was developed by which all normal distributions can be converted into a single
distribution: the 𝑧 distribution. This process yields the standardized normal distribution (or
curve). The conversion formula for any 𝑥 value of a given normal distribution follows.
𝑥−𝜇
𝑧= , 𝜎≠0
𝜎
A 𝒛 score is the number of standard deviations that a value, x, is above or below the mean.
If the value of 𝑥 is less than the mean, the 𝑧 score is negative; if the value of 𝑥 is more than
the mean, the z score is positive; and if the value of 𝑥 equals the mean, the associated 𝑧
score is zero.
This formula allows conversion of the distance of any 𝑥 value from its mean into standard
deviation units. A standard 𝑧 score table can be used to find probabilities for any normal
curve problem that has been converted to 𝑧 scores.
If we take several random samples from a normally distributed population of observations, the
means of such samples will also be normally distributed around the population mean. Therefore,
the probability that a randomly selected sample will have a mean outside of the critical region is
the same the probability of rejecting a null hypothesis that is true (Type I error).
Null and Alternate Hypothesis
All statistical hypotheses consist of two parts, a null hypothesis and an alternative hypothesis.
These two parts are constructed to contain all possible outcomes of the experiment or study.
Generally, the null hypothesis states that the “null” condition exists; that is, there is
nothing new happening, the old theory is still true, the old standard is correct, and the system
is in control.
The alternative hypothesis, on the other hand, states that the new theory is true, there
are new standards, the system is out of control, and/or something is happening.
The null and alternative hypotheses for the flour example can be restated as:
𝐻0 : 𝜇 = 40
𝐻0 : 𝜇 ≠ 40
suppose a company has held an 18% share of the market. However, because of an increased
marketing effort, company officials believe the company’s market share is now greater than 18%,
and the officials would like to prove it. The null hypothesis is that the market share is still 18% or
perhaps it has even dropped below 18%. Converting the 18% to a proportion and using p to
represent the population proportion, results in the following hypotheses:
𝐻0 : 𝑝 ≤ 0.18
𝐻0 : 𝑝 > 0.18
One-tailed tests are always directional, and the alternative hypothesis uses either the greater
than (>) or the less than (<) sign. A one-tailed test should only be used when the researcher
knows for certain that the outcome of an experiment is going to occur only in one direction or the
researcher is only interested in one direction of the experiment as in the case of the market share
problem.
CORRELATION AND REGRESSION
Sometimes, the forecaster will wish to predict one variable 𝑌 (e.g., Sales) and have
available one explanatory variable 𝑋 (e.g., advertising expenditure). The objective is to
develop an explanatory model relating 𝑋 and 𝑌. This is called simple regression.
If there is one variable to forecast (𝑌) and several explanatory variables (𝑋1 , 𝑋2 … 𝑋𝐾 ) and
the objective is to find a function that relates 𝑌 to all of the explanatory variables, this is
called multiple regression of 𝑌 on 𝑋1 through 𝑋𝐾 .
If the data are measured over time, then it will be called time series regression.
If the data measurements are taken all at the same time, it will be called a cross-sectional
regression.