Regression
5-Jan-20
Great Learning
1
Agenda
1.
Introduction to Linear regression
2.
Correlation vs regression
3.
Applications of Linear regression
4.
Assumptions of Linear regression
5.
Two main types of Linear regression
6.
Simple Linear regression (SLR) example
7.
Multiple Linear regression (MLR) example

Regression
5-Jan-20
Great Learning
2
1.
Introduction to Linear regression
a.
Sir Francis Galton, a British statistician coined the term
regression based on
his research
on hereditary properties of successive generations of sweet peas
and humans. In 1886, he published an article with title,
Regression towards
Mediocrity in Hereditary Stature
in Journal of the anthropological institute of
Great Brittan and Ireland).
b.
Linear regression is a mathematical technique to establish a relationship
between two variables (predictor variable and response variable) by finding a
straight line that best fits the values of a linear function.

Regression
5-Jan-20
Great Learning
3
1.
Introduction to Linear regression
c.
Scatter plots gives us an idea whether the two variables are linearly related or
not.
We need to find the best line that represents the scatter using linear
regression.
d.
Response variable (aka dependent or outcome or target) is the variable of
focus in a research study.
e.
Predictor variable (aka independent or explanatory) is the variable that
explains the variation in the response variable and it might affect the
response variable.

Regression
5-Jan-20
Great Learning
4
1.
Introduction to Linear regression - continued
f.
For example, the response variable is volume of
Sales in thousands on a given day in an online
stores
and
the
predictor
variable
is
the
advertisement expenses.
g.
Focus
of
the
regression
analysis
is
on
the
relationship between a response variable and
one or more predictor variables. To be more
specific, this helps one to understand how the
typical value of the response variable changes
when
any
one
of
the
predictor
variables
is
varies,
keeping
other
predictor
variables
constant.

Regression
5-Jan-20
Great Learning
5
1.
Introduction to Linear regression - continued
h.
Identification of problem
Before doing the regression analysis, as a data scientist you must review the
relevant literatures to develop a deep understanding of the business
domain
to know the relevant variables, their relationships.
The predictor (independent) variable is the core of the experiment and is isolated and
manipulated by the researcher. A researcher must determine which variable is
reliable and relevant that needs to be manipulated to generate quantifiable results.
For more details. refer
https
://explorable.com/research-variables/

Regression
5-Jan-20
Great Learning
6
2.
Correlation vs regression
✓
Correlation is a measure and direction of strength of a linear relationship
between two variables. The Pearson's correlation coefficient or correlation
coefficient which is valid only for linear relationship, denoted by r is a value
that ranges between -1 and 1; -1 indicates perfect negative relationship, 1
indicates perfect positive relationship and 0 indicates no relationship.

#### You've reached the end of your free preview.

Want to read all 66 pages?

- Fall '19
- Regression Analysis