DSILYTC:

Introduction to
Analytics
SESSION 5: DESCRIPTIVE STATISTICS
Objective of the Study

 DescriptiveAnalytics
 Case Analysis:
 Applications of
Numerical Measures
DESCRIPTIVE STATISTIC: NUMERICAL MEASURES

 MEASURE OF LOCATION • MEASURE OF VARIABLES

 NUMERICAL MEASUREMENT

 If
the measures are computed for data from
a sample, they are called SAMPLE STATISTIC.
 Ifthe measures are computed for data form
a population, they are called POPULATION
PARAMETERS.
A sample statistic is referred to as the point
estimator of the corresponding population
parameter.
MEASURE OF LOCATION

 Mean
 Median
 Mode
 Weighted Mean
 Geometric Mean
 Percentile
 Quartiles
 MEAN

 Perhaps the most important measure of location

is the mean
 The mean provides a measure of central location
 The mean of a data set is the average of all the data
values
 Thesample mean is the point estimator of the
population means
 MEDIAN

 The median of a data set is the value in the middle when

the data items are arranged in ascending order.
 Whenever a data set has extreme values, median is the
preferred measure of central location.
 The median is the measure of location most often
reported for annual income and property value data.
 A few extremely large income or property values can inflate
the mean.
MEDIAN
MEDIAN
MEDIAN
 TRIMMED MEAN

 Another measure sometimes used when extreme

values are present is the TRIMMED MEAN.
 It is obtained by deleting a percentage of the
smallest and largest values from a data set and
then computing the mean of the remaining
values.
 For example, the 5% trimmed mean is obtained by
removing the smallest 5% and the largest 5% of the
data values and then computing the mean of the
remaining values.
 MODE

 The mode of a data set is the value that

occurs with greatest frequency.
 The greatest frequency can occur at two
or more different values.
 If the data have exactly 2 modes, the
data are BIMODAL.
 If the data have more than 2 modes, the
data are MULTIMODAL.
MODE
 WEIGHT MEAN

 In some instance the mean is computed by

giving each observation a weight that
reflects its relative importance.
 The choice of weights depends on the
applications.
 The weight might be the number of credit
hours earned for each grade, as in GPA
 In other weighted mean computations,
quantities such as pounds, dollars, or volume
are frequently used.
WEIGHT MEAN
WEIGHT MEAN
WEIGHT MEAN
 GEOMETRIC MEAN

 The geometric mean is calculated by findings the

nth root of the product of n values.
 It is often used in analyzing growth rates in financial
data (where using the arithmetic mean will provide
misleading results).
 It should be applied anytime you want to determine
the mean rate of change over several successive
periods (be it years, quarters, weeks,…)
 Other common applications include: change in
population of species, crop yields, pollution levels,
and birth and death rates.
GEOMETRIC MEAN
GEOMETRIC MEAN
 PERCENTILES

 A percentiles provides information about how the

data are spread over the interval from the smallest
value to the largest value.
 Admission test scores for colleges and universities
are frequently reported in terms of percentiles.
 The pth percentiles of a data set is value such that
at least p percent of the items takes on this value
or less and at least (100 – p) percent of the items
take on this value or more.
PERCENTILES
QUARTILES
 MEASURES OF VARIABILITY

 Itis often desirable to consider measures

of variability (dispersion), as well as
measures of location.
 Forexample, in choosing supplier A or
supplier B we might consider not only the
average delivery time for each, but also
the variability in delivery time for each.
 MEASURES OF VARIABILITY

 Range
 Interquartile Range
 Variance
 Standard deviation
 Coefficient of variation
 RANGE

 The range of data set is the difference between the largest and
smallest data values.

RANGE = LARGEST Values -

SMALLEST Values

 It is the simplest measure of variability

 It is very sensitive to the smallest and largest data values.
 INTERQUARTILE RANGE

 The interquartile range of a data set is

the difference between the 3rd
quartile and the 1st quartile.
 It is the range for the middle 50% of
the data.
 It overcomes the sensitivity to extreme
data values.
INTERQUARTILE RANGE
 VARIANCE

 The variance is a measure of variability that

utilize all the data.
 It is based on the difference between the value
of each observation

 The variance is useful in comparing the

variability of 2 or more variables
VARIANCE
 STANDARD DEVIATION

 The standard deviation of a data set

is the positive square root of the
variance.
 It is measured in the same units as the
data, making it more easily
interpreted than the variance.
STANDARD DEVIATION
 COEFFICIENT OF VARIATION

 Thecoefficient of variation indicates

how large the standard deviation is in
relation to the mean.
 The coefficient of variation is computed as
follows:
Descriptive Statistic: Numerical
Measures (part 2)
 Measure of Distribution shape, Relative
Location and Detecting Outliers.
 Five number summaries and Box plots
 Measures of Association between 2
variables
 DataDashboard: Adding numerical
measures to improve effectiveness
Measure of Distribution shape, Relative
Location and Detecting Outliers.

 Distribution shape
 Z-scores
 Chebyshev’s Theorem
 Empirical Rule
 Detecting Outliers
Distribution shape

 An important measure of the shape of a

distribution is called SKEWNESS.
 The formula for the skewness of sample data is

 The skewness can be easily computed using

statistical software.
Distribution shape: SKEWNESS
Distribution shape: SKEWNESS
Distribution shape: SKEWNESS
Distribution shape: SKEWNESS
 Z-Scores

 The z-scores is often called the standardized

value.
 It denotes the number of standard deviations a
data value is form the mean.

 Excel’s STANDARDIZE function can be used to

compute the z-scores.
 Z-Scores

 An observations’ z-scores is a measure of the

relative location of observation in a data set.
 A data value less than the sample mean will
have a z-score less than zero.
 A data value greater than the sample mean will
have a z-score greater than zero.
 A data value equal to the sample mean will
have a z-score of zero.
Z-Scores
Chebyshev’s Theorem
 Chebyshev’s Theorem

 At least 75% of the data values must be

within z= 2 standard deviation of the
mean.
 At least 89% of the data values must be
within z=3 standard deviation of the
mean.
 At least 94% of the data values must be
within z=4 standard deviation of the
mean.
Chebyshev’s Theorem
Chebyshev’s Theorem
 EMPIRICAL RULE

 Whenthe data are believed to

approximate a bell-shaped distribution:
 Theempirical rule can be used to
determine the percentage of data values
that must be within a specified number of
standard deviations of the mean.
 The empirical rule is based on the normal
distribution, which is covered in chapter 6.
 EMPIRICAL RULE

For data having bell-shaped distribution.

Approximately 68% of the data values will be
within +/-1 standard deviation of its mean.
Approximately 95% of the data values will be
within +/-2 standard deviation of its means.
Almost all of the data values will be within +/-3
standard deviation of its means.
EMPIRICAL RULE
 Detecting Outliers

 An outliers is an unusually small or unusually large

value in a data set.
 A data value with a z-score less than -3 or greater
than +3 might be considered an outlier.
 It might be
 An incorrectly recorded data value
 A data value that was incorrectly included in the
data set
 A correctly recorded unusual data value that belongs
in the data set.
OUTLIERS
FIVE NUMBER SUMMARIES & BOX
PLOT
 Summary statistic and easy-to-draw
graphs can be used to quickly
summarize large quantities of data.
 Two tools that accomplish this are
five-number summaries and box plots.
FIVE NUMBER SUMMARY

 Smallest value
 First quartile
 Median
 Third quartile
 Largest Value
FIVE NUMBER SUMMARY
 BOX PLOT

A box plot is a graphical summary of data

that is based on a five-number summary
A key to the development of a box plot is
the computation of median and quartiles
Q1 and Q3
 Box plot provide another way to identify
outliers.
BOX PLOT
 BOX PLOT

 Limitsare located (not drawn) using the

interquartile range (IQR).
 Data outside these limits are considered
(outliers).
 The location of each outlier is shown with
the symbols.
BOX PLOT
BOX PLOT
Measures of Association between 2
variables
 Thus far we have examined numerical
methods used to summarize the data for one
variables at a time.
 Often a manager or decision maker is
interested in the relationship between 2
variables.
 Two descriptive measures of the relationship
between 2 variables COVARIANCE and
CORRELATION COEFFICIENT.
 COVARIANCE

 The covariance is a measure of the

linear association between 2
variables.
 Positive values indicates a positive
relationship.
 Negative values indicates a
negative relationship.
COVARIANCE
 CORRELATION COEFFICIENT

Correlation is a measure of linear

association and not necessarily
causation
Just between 2 variables are highly
correlated, it does not mean that one
variables is the cause of the other.
CORRELATION COEFFICIENT
 CORRELATION COEFFICIENT

 The coefficient can take on value

between -1 and +1
 Values near -1 indicate a strong negative
linear relationship.
 Values near +1 indicate a strong positive
linear relationship
 The closer the correlation to zero, the
weaker relationship.
CORRELATION COEFFICIENT
COVARIANCE & CORRELATION
COEFFICIENT
COVARIANCE & CORRELATION
COEFFICIENT
COVARIANCE & CORRELATION
COEFFICIENT
 DATA DASHBOARDS:
Adding numerical measure to
 Data improve effectiveness
dashboard are not limited to graphical
displays
 The addition of numerical measures, such as
the mean and standard deviation of KPI’s, to
a data dashboard is often critical.
 Dashboards are often interactive.
 Drilling Down refers to functionally in
interactive dashboards that allows the users
to access information and analyses at
increasingly detailed level.
 DATA DASHBOARDS:
Adding numerical measure to
improve effectiveness
DSILYTC:
Introduction to
Analytics
SESSION 5: DESCRIPTIVE STATISTICS