Nonparametric Statistics:
All of the hypothesis testing methods we have covered up to this
point assume that the data follows a certain distribution, generally the
normal distribution.
These are considered to be
parametric
methods
.
In real life, data sometimes fails to meet this assumption.
Several
methods have been developed to test data without assuming any
knowledge of the distribution of the underlying populations, except
that they are continuous.
These methods are called
nonparametric
or distributionfree methods
.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentONE SAMPLE TESTS
1) Sign Test:
This method is used when we want to test Ho:
0
~
~
μ
=
and n<30 and
the population is not normal.
Assumption:
X
1
, X
2
, X
3,
…
,
X
n
are independently and identically
distributed.
Our hypotheses can be onesided or twosided:
Ho:
0
~
~
=
H1:
0
~
~
or
0
~
~
μ<
or
0
~
~
μ≠
Calculating the Test Statistic:
Find D
i
=X
i

μ
0
for each sample value. If we get any D
i
=0, remove
that from the sample and decrease the sample size by 1 for each D
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '07
 MELutz
 Statistics, Normal Distribution, Statistical hypothesis testing, Nonparametric statistics, hypothesis testing methods

Click to edit the document details