### Statistics Phase 5 : Correlation Vs Causation

__COVARIANCE__
Covariance
is a measure of the relationship between 2 or more random variables or how 2
random variables vary together. Its similar to variance, but where variance
tells you how a single variable varies, covariance tells you how 2 variables
vary together.

Positive
covariance means as one variable increases the other one also increases. For
example: lets’ take 2 variables height and weight. As heights increases weight
also increases.

Negative
covariance means as one decreases, the other one also decreases. For example,
as salary decreases, the expenditure also decreases.

Measuring
something in inches would be say 12 and converting the same into centimetres
would be different only because of the unit change. So its hard to tell how
strong the relationship is based on the actual magnitude of the covariance.

__CORRELATION__
On one
hand, covariance indicates the direction of the linear relationship between the
variables whereas correlation on the other hand, indicates both the strength
and the direction of the linear relationship.

When
correlation coefficient is 0, covariance is also 0.

For
example: as the number of accident claims per month increase, the number of
individuals asking for the claim decreases or as temperature increases, the
demand for coolers increases.

ADVANTAGES
OF CORRELATION OVER COVARIANCE

·
Covariance
can take any number while correlation is limited; -1 to +1

·
Correlation
is more useful for determining how strong the relationship is between 2
variables.

·
Correlation
does not have units unlike covariance.

·
Correlation
remains unaffected by the change in scale.

Due to
all these reasons, correlation is preferred over covariance.

__CORRELATION VS CAUSATION__
A
statistical measure (expressed as a number) that describes the size and
direction of a relationship between two or more variables is correlation
whereas while causation indicates that one event is the result of the occurrence
of the other event; i.e. there is a causal relationship between the two events.
This is also referred to as cause and effect.

The
causation vs correlation example that is frequently used is that smoking is
correlated with alcoholism, but doesn’t cause alcoholism. While smoking causes
an increase in the risk of developing lung cancer.

When
a person is exercising then the amount of calories burning goes up every
minute. Former is causing latter to happen.

**Follow us on LinkedIn : Actuary Sense**

**Follow me on LinkedIn: Kamal Sardana**

**credits: Simran Agrawal**
This comment has been removed by the author.

ReplyDeleteYour explanation related to negative covariance is wrong.

ReplyDeleteIf both the variables move in same direction it is positive covariance and in negative covariance both the variable move in opposite direct.

Example for negative covariance is

Price and demand

If price increases demand decreases and vice versa.

yeah raj you are right

Delete