Quantitative Methods: Sessions 1-3 Case: Catalog Marketing
Quantitative Methods: Sessions 1-3 Case: Catalog Marketing
SESSIONS 1-3
CASE:
Catalog Marketing
CONNECT!
Poll
http://timesofindia.indiatimes.com/home/polls
http://www.economist.com/news/finance-and-economics/21698656-jacking-up-prices-may-
not-be-only-way-balance-supply-and-demand-taxis?fsrc=scn/fb/te/bl/ed/
Some useful databases
• data.gov.in – This is the portal of the Indian Government’s open data. You can
check out a few visualizations for inspiration here.
• World Bank – The open data from the World bank. The platform provides several tools
like Open Data Catalog, world development indices, education indices etc.
• Five Thirty Eight Datasets – Here is a link to datasets used in different stories. Each
dataset includes the data, a dictionary explaining the data and the link to the story
carried out by Five Thirty Eight.
COURSE EVALUATION
Categorical Numerical
Examples:
Marital Status
Political Party Discrete Continuous
Eye Color
(Defined categories) Examples: Examples:
Number of Children Weight
Defects per hour Voltage
(Counted items) (Measured characteristics)
Four Levels of
Measurement
Nominal level – data that is Interval level – similar to the
classified into categories and ordinal level, with the additional
cannot be arranged in any property that meaningful amounts of
particular order. differences between data values can
be determined. There is no natural
zero point.
Ordinal level – data arranged in Ratio level – the interval level with
some order, but the differences an inherent zero starting point.
between data values cannot be Differences and ratios are meaningful
determined or are meaningless. for this level of measurement.
1-11
Data Scales
Scale Basic Common Marketing Permissible Statistics
Characteristics Examples Examples Descriptive Inferential
Nominal Numbers identify Social Security Brand nos., store Percentages, Chi-square,
& classify objects nos., numbering types mode binomial test
of football players
Ordinal Nos. indicate the Quality rankings, Preference Percentile, Rank-order
relative positions rankings of teams rankings, market median correlation,
of objects but not in a tournament position, social Friedman ANOVA
the magnitude of class
differences
between them
Interval Differences Temperature Attitudes, Range, mean, Product-moment
between objects (Fahrenheit) opinions, index standard correlation, t
can be compared, Celsius) nos. deviation tests, regression
zero point is
arbitrary
Ratio Zero point is Length, weight Age, sales, Geometric mean, Coefficient of
fixed, ratios of income, costs harmonic mean variation
scale values can
be compared
Descriptive Statistics
Statistic is a value that summarizes the data of a particular variable.
The central tendency is the extent to which all the data values group around a typical or
central value.
The shape is the pattern of the distribution of values from the lowest value to the highest
value.
Mean
The arithmetic mean (often just called the “mean”) is the most common
measure of central tendency
X i
X1 X2 Xn
X i1
n n
Sample size Observed values
Mean
The population mean is the sum of the values in the population divided by
the population size, N
X i
X1 X2 XN
i1
N N
Where μ = population mean
N = population size
Xi = ith value of the variable X
MEAN
The most common measure of central tendency
Mean = sum of values divided by the number of values
Affected by extreme values (outliers)
11 12 13 14 15 16 17 18 19 20 11 12 13 14 15 16 17 18 19 20
Mean = 13 Mean = 14
11 12 13 14 15 65 11 12 13 14 20 70
13 14
5 5 5 5
Median
In an ordered array, the median is the “middle” number (50% above, 50%
below)
11 12 13 14 15 16 17 18 19 20 11 12 13 14 15 16 17 18 19 20
Median = 13 Median = 13
Not affected by extreme values
Median
The position of the median when the values are in numerical order (smallest to largest):
n 1
Median position position in the ordered data
2
If the number of values is odd, the median is the middle number
If the number of values is even, the median is the average of the two middle numbers
n 1
2
Mode
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 0 1 2 3 4 5 6
No Mode
Mode = 9
Measures of Central Tendency:
Review Example
House Prices: Mean: ($3,000,000/5)
$2,000,000 = $600,000
$ 500,000
$ 300,000
Median: middle value of ranked
$ 100,000 data
$ 100,000 = $300,000
Sum $ 3,000,000 Mode: most frequent value
= $100,000
Measures of Central Tendency:
Which Measure to Choose?
X i
X i1
n Middle value Most
in the ordered frequently
array observed
value
Measures of Variation
Variation
Example:
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Range = 13 - 1 = 12
Why The Range Can Be Misleading
• Ignores the way in which data are distributed
7 8 9 10 11 12 7 8 9 10 11 12
Range = 12 - 7 = 5 Range = 12 - 7 = 5
• Sensitive to outliers
1,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,3,3,3,3,4,5
Range = 5 - 1 = 4
1,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,3,3,3,3,4,120
Range = 120 - 1 = 119
Variance
Average (approximately) of squared deviations of values from the mean
( xi )
N 2 n
(X X)
2
2 i
i 1 N S 2 i1
n -1
X = Sample mean
Where
n = sample size; N = population size
Xi = ith value of the variable X
= Population Mean
Standard Deviation
Most commonly used measure of variation
Shows variation about the mean
Is the square root of the variance
Has the same units as the original data
( xi )
N 2
(X X)
i
2
N S i1
i 1 n -1
Measures of Variation:
Comparing Standard Deviations
Smaller standard deviation
Sample
Data (Xi) : 10 12 14 15 17 18 18 24
n=8 Mean = X = 16
Data A
Mean = 15.5
11 12 13 14 15 16 17 18 19 20 21 S = 3.338
Q1 Q2 Q3
The first quartile, Q1, is the value for which 25% of the
values are smaller and 75% are larger.
Q2 is the same as the median (50% of the values are
smaller and 50% are larger).
Only 25% of the values are greater than the third quartile.
Quartile Measures:
Locating Quartiles
◦ If the result is a fractional half (e.g. 2.5, 7.5, 8.5, etc.) then average the two corresponding data values.
◦ If the result is not a whole number or a fractional half then round the result to the nearest integer to
find the ranked position.
Quartile Measures:
Locating Quartiles
Sample Data in Ordered Array: 11 12 13 16 16 17 18 21 22
(n = 9)
Q1 is in the (9+1)/4 = 2.5 position of the ranked data
so use the value half way between the 2nd and 3rd values,
so Q1 = 12.5
(n = 9)
Q1 is in the (9+1)/4 = 2.5 position of the ranked data,
so Q1 = (12+13)/2 = 12.5.
Example:
Median X
X Q1 Q3 maximum
minimum (Q2)
25% 25% 25% 25%
12 30 45 57 70
Interquartile range
= 57 – 30 = 27
Five Number Summary and
The Boxplot
The Boxplot: A Graphical display of the data based on the five-number
summary:
Q1 Q2 Q3 Q1 Q2 Q3 Q1 Q2 Q 3
Measures of Variation:
Summary Characteristics
• The more the data are spread out, the greater the range, variance, and
standard deviation.
• The more the data are concentrated, the smaller the range, variance, and
standard deviation.
• If the values are all the same (no variation), all these measures will be
zero.
Can be used to compare the variability of two or more sets of data measured
in different units
S
CV 100%
X
Measures of Variation:
Comparing Coefficients of Variation
Stock A:
◦ Average price last year = $50
◦ Standard deviation = $5
S $5
CVA 100% 100% 10%
X $50 Both stocks
have the same
standard
Stock B: deviation, but
◦ Average price last year = $100 stock B is less
variable relative
◦ Standard deviation = $5 to its price
S $5
CVB 100% 100% 5%
X $100
Measures of Variation:
Comparing Coefficients of Variation
Stock A:
◦ Average price last year = $50
◦ Standard deviation = $5
S $5
CVA 100% 100% 10%
X $50 Stock C has a
much smaller
Stock C: standard
deviation but a
◦ Average price last year = $8 much higher
◦ Standard deviation = $2 coefficient of
variation
S $2
CVC 100% 100% 25%
X $8
Shape of a Distribution
Describes how data are distributed
Two useful shape related statistics are:
◦ Skewness
◦ Measures the extent to which data values are not symmetrical
◦ Kurtosis
◦ Kurtosis affects the peakedness of the curve of the distribution—that is, how
sharply the curve rises approaching the center of the distribution
Shape of a Distribution (Skewness)
Measures the extent to which data is not symmetrical
Skewness
Statistic <0 0 >0
Coefficient of Skewness
Summary measure for skewness
3 Md
Sk
Coefficient of Skewness (Sk) - compares the mean
and median in light of the magnitude to the standard deviation;
Md is the median; Sk is coefficient of skewness; σ is the Std Dev
4-48
Step 1 : Compute the Mean
X
X
$74.26
$4.95
n 15
s
X X
2
($0.09 $4.95) 2 ... ($16.40 $4.95) 2 )
$5.22
n 1 15 1
4-49
Sample Skewness
Step 4 : Compute sk
3( X M )
s
3 * (4.95 3.18)
5.22
1.017241
However, when our data is skewed, for example, as with
the right-skewed data set given here.
(x x)i
4
kurtosis i
4
ns
The kurtosis of a normal distribution is 3
Kurtosis characterizes the relative peakedness or flatness or
tailedness of a distribution compared to the normal distribution
Shape of a Distribution -- Kurtosis
Sharper Peak
Than Bell-Shaped
(Kurtosis > 0)
Bell-Shaped
(Kurtosis = 0)
Flatter Than
Bell-Shaped
https://upload.wikimedia.org/wikipedia/commons/e/e6/Stand (Kurtosis < 0)
ard_symmetric_pdfs.png
Mean=0; Standard Deviation=1; Skewness=0
f * M 2150
43.0
f 50
Median of Grouped Data - Example
Select the class interval with cumulative frequency just greater than N/2
Cumulative
Class Interval Frequency Frequency
20-under 30 6 6
N
30-under 40 18 24 cfp
40-under 50 11 35 Md L 2 W
fmed
50-under 60 11 46
50
60-under 70 3 49 24
70-under 80 1 50 40 2 10
11
N = 50 40.909
L: Lower limit of the median class
cfp : cumulative frequency of the previous class to the median class
fmed : frequency of median class
W: Length of the class interval
Mode of Grouped Data
Midpoint of the modal class
Modal class has the greatest frequency
f M M X
2 2
f
2
2
N S n1
2
S
2
S
Regardless of how the data are distributed, at least (1 - 1/k2) x 100% of the
values will fall within k standard deviations of the mean (for k > 1)
◦ Examples:
μ 1σ
68%
μ
μ 1σ
The Empirical Rule
95% 99.7%
μ 2σ μ 3σ
Using the Empirical Rule
Suppose that the variable Math SAT scores is bell-shaped with a mean of 500 and a standard
deviation of 90. Then,
68% of all test takers scored between 410 and 590 (500 ± 90).
95% of all test takers scored between 320 and 680 (500 ± 180).
99.7% of all test takers scored between 230 and 770 (500 ± 270).
The Covariance
The covariance measures the strength of the linear relationship
between two numerical variables (X & Y).
( X X)(Y Y)
i i
cov ( X, Y) i1
n 1
cov (X, Y)
r
SX SY
Where,
(X X)(Y Y)
n n
i i (X X)
i
2
(Y Y)
i
2
X X
r = -1 r = -.6
Y
Y Y
X X X
r = +1 r = +.3 r=0
Calculate Mean, Median, mode, standard deviation, coefficient of
variation for a sample of usage times of 50 ATM customers.
Time Frequency
20-25 1
25-30 7
30-35 10
35-40 9
40-45 9
45-50 6
50-55 5
55-60 3