This document discusses key concepts in descriptive statistics including measures of central tendency, spread, skewness, kurtosis, and correlation. It covers the stages of analytics from descriptive to predictive, the data life cycle process, data types, and calculating skewness and kurtosis coefficients. Correlation is defined as the analysis of co-variation between variables and the difference between correlation and causation is explained.

1. Descriptive
Statistics
Skewness
& Kurtosis
Correlation

2. Agenda
•Stages of Analytics
•Data Life Cycle
•Data Types
•First Business Moment Decision (Measure of CentralTendency)
•Second Business Moment Decision (Measure of Spread)
•Third Business Moment Decision (Skewness)
•Fourth Business Moment Decision(Kurtosis)
•Correlation

3. CRISP
DM
Business
Understandi
ng
Data
Acquisition
Data
Processing
Data
Exploration
Modelling
Model
Evaluation
Deployment
Data Life Cycle

4. 1. Descriptive Analytics
2. Diagnostic Analytics
3. Predictive analytics
4. Prescriptive Analytics
Stages of Analytics

5. Data
Quantitative
Continuous
Interval Ratio
Discrete
Qualitative
Categorical
Nominal Ordinal

9. Balanced Vs Imbalanced Data

26. Skewness
degree of asymmetry observed in a probability distribution that deviates from the symmetrical normal distribution
(bell Curve)

27. 1. Positive skewed or right-skewed
The extreme positive skewness is not
desirable for distribution, as a high level of
skewness can cause misleading results.
For positively skewed distributions, the
famous transformation is the log
transformation. The log transformation
proposes the calculations of the natural
logarithm for each value in the dataset.

28. 2. Negative skewed or left-skewed
a large number of data-pushed on the left-
hand side
Median is the middle value, and mode is the
highest value, and due to unbalanced
distribution median will be higher than the
mean.

29. •If the skewness is between -0.5 & 0.5, the data are nearly symmetrical.
•If the skewness is between -1 & -0.5 (negative skewed) or between 0.5 & 1(positive skewed), the
data are slightly skewed.
•If the skewness is lower than -1 (negative skewed) or greater than 1 (positive skewed), the data are
extremely skewed.
Calculate the skewness coefficient of the sample

30. Kurtosis
Kurtosis refers to the degree of
presence of outliers in the
distribution.
Kurtosis is a statistical measure,
whether the data is heavy-tailed or
light-tailed in a normal distribution

31. Types of Kurtosis
1.Leptokurtic (kurtosis > 3)
2.platykurtic (kurtosis < 3)
3.Mesokurtic (kurtosis = 3)

32. Correlation
1.“Correlation is an analysis of the co-variation between two or
more variables”—(A.M Tuttle)
2. “Correlation analysis attempts to determine the degree of
relationship between variables”—(Ya Lun Chou)
3. “Correlation analysis deals with the association between two
or more variables”— (Simpson and Kafka)

33. Correlation and Causation
Correlation: It is a numerical measure of the direction
and magnitude of the mutual relationship between the
variables(X and Y).
Causation: X is the cause of change in Y i.e, the
change of Y is the effect of change in X.
NOTE:
If X and Y are correlated then X and Y may or may not have a casual relationship.
If X and Y have a causal relationship then X and Y must be correlated.

34. Reasons Behind Correlation
1. Mutual dependence Between the variables: Both the variables may be mutually
influencing each other so that neither can be designated as the cause and the other the
effect.
When two variables(X and Y) affect each other mutually, we cannot say X is the cause or Y
is the cause.
For Example, The price of a commodity is affected by demand and supply.
2. Due to pure chance: In a small sample, X and Y are highly correlated but in the
universe X and Y are not correlated.
For Example, Correlation between income and weight of a person. This may be due to:
– Sampling fluctuations
– Bias of investigator in selecting the sample.
Such a relation is called a non-sense or spurious relation.
3. Correlation due to any third common factor: Both the correlated variables may be
influenced by one or other variables.
– X and Y don’t have a direct correlation.
For Example, It is between the production of tea and rice per hectare. Here they are not
directly correlated instead the cause is the good rainfall well in time.

35. Types of Correlation

36. Key Takeaway
After this session you learnt:
1. Stages of Analytics
2. Data Life Cycle
3. Data Types
4. First Business Moment Decision (Measure of CentralTendency)
5. Second Business Moment Decision (Measure of Spread)
6. Third Business Moment Decision (Skewness)
7. Fourth Business Moment Decision(Kurtosis)
8. Correlation