B ASIC S TATISTICAL C ONCEPTS F OR

S USTAINABLE M ANUFACTURING
A NALYSIS

Dr. Deepu Philip

1 8/26/2017
2 A GENDA

Random Variables
Independence & Randomness
Covariance Stationary Process
Law of Large Numbers
Probability Distributions & Means
Test of Independence

8/26/2017
3 B ASIC T ERMINOLOGIES

Experiment
Process whose outcome is not known with certainty

Sample space
Collection of all possible outcomes of experiments

Sample points
Outcomes are sample points in sample space

8/26/2017
4 R ANDOM VARIABLES

Random Variable is a function that assigns a

real number to each point in sample space
Probability density function or mass function
Discrete
Continuous

Cumulative density function or distribution

function
Discrete
Continuous
8/26/2017
5 M EAN & VARIANCE

Mean or Expected value of a random variable

Mean provides the central tendency
Variance of a random variable measures it
dispersion
Variance is always greater than or equal to zero

Standard deviation
Has the same unit as that of the data

8/26/2017
6 I NDEPENDENCE
Two events are independent if the occurrence of
one event does not affect the probability of
occurrence of the other
P(A/B) = P(B)

Independent events must be able to occur at the

same time
Disjoint events cannot be independent
Disjoint events cannot happen at the same time

Not disjoint events may or may note be

independent
8/26/2017
7 I LLUSTRATIVE E XAMPLES

Draw a card from deck and bet on whether the

color is red
Before placing bet, it is told that the card is spade
red and spade are disjoint events, but that
information influence the event bet.

Before placing bet, it is told that card is ace ace

and red are not disjoint, but they are independent.
P(Red) = 26/52 =
P(Red/Ace) = 2/4 =
The occurrence of ace has no influence on red
8/26/2017
8 A NOTHER E XAMPLE

Colorblindness occur more often in men than in

woman
But a woman can also be colorblind, hence the
events are not disjoint
Given that the probability of woman being
colorblind is much lower, sex and
colorblindness are not independent either.

8/26/2017
9 M EASURE O F D EPENDENCE

Covariance measures the linear dependence of

two random variables
Cov(X, Y) = E(XY) E(X)E(Y)
If Cov(X, Y) = 0 then the random variables are
uncorrelated.
If X and Y are independent, then Cov(X, Y) = 0
Random variables can be positively or
negatively correlated
8/26/2017