EDA - INFERENTIAL STATISTICS

One Sample Hypothesis testing

tests an assumption about the

population based on the sample statistic, giving us the following null

and alternative hypothesis formats:

Two sample Hypothesis testing

compares the population parameter of

two sets of data, giving us the following null and alternative hypothesis

formats, with d0 (or µD, dµ, or pD, dp from other sources) being the

assumed difference of the said population parameters.

Independent testing

If sample set A has no relation to sample set B

• If the entities in sample set A are different from the entities of

sample set B

• If sample set A is not an effect of sample set B, vice versa

Dependent Testing (Matched Pairs)

If sample set A has immediate relation to sample set B

• If the entities in sample Set A are the same entities in sample

set B

• If sample set A is an effect of sample set B, vice versa

• If the sample sets are paired

MULTIPLE

CORRELATION

yields the maximum degree of linear

relationship that can be obtained between two or more independent

variables and a single dependent variable.

INPUT (INDEPENDENT VARIABLES)

Materials, information, labor, energy, conditions

PROCESS (SYSTEM)

Machine, chemical reaction, human activity,

program

OUTPUT (DEPENDENT VARIABLE)

Products, results, possible outcomes