Scribd
Upload
9 views2 pages

Statistical Tests Overview and Methods

The document outlines various statistical tests, including Z-test, T-test, Poisson test, Binomial test, and tests for independence and goodness of fit (GOF). Each test is associated with its purpose, null and alternative hypotheses, prerequisites, and relevant statistical methods. Additionally, it discusses the concepts of Type 1 and Type 2 errors in hypothesis testing.

Uploaded by

tsinapah
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views2 pages

Statistical Tests Overview and Methods

The document outlines various statistical tests, including Z-test, T-test, Poisson test, Binomial test, and tests for independence and goodness of fit (GOF). Each test is associated with its purpose, null and alternative hypotheses, prerequisites, and relevant statistical methods. Additionally, it discusses the concepts of Type 1 and Type 2 errors in hypothesis testing.

Uploaded by

tsinapah
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Assumptions/

Test Purpose H0 H1 statistic test Null distribution GDC


prerequisites
- known Z-Test
Z-test Test if the - data normally normcdf
population mean distributed invNorm
of a equals a - unknown
value
T-test - data normally GDC GDC T-Test
distributed
Compare the
2 sample T- means of two data normally
GDC GDC 2-SampleTTest
test unpaired independent distributed
samples

Compare the T-Test on the


2 sample T- data normally
means of two GDC GDC difference of
test paired distributed
related samples. the data

Test if the - data Poisson


poissoncdf
population mean distributed
Poisson test find c with
of a equals a - Observations are
value
iteration
independent.
- Sample size is
Is the sample
large enough
proportion of a
characteristic ( and binomcdf
t = number of
Binomial test equal to the . if p not given: find c with
success in the sample
population iteration
- Observations are (estimator)
proportion.
independent.

To test if there
is a correlation data follows a
Population
between two bivariate normal GDC GDC LineReg-TTest
Correlation
variables in a distribution
population
Assumptions/ GDC
Test Purpose H0 H1 degree of freedom expected frequencies
prerequisites (p-value, c)
To determine if
two categorical -Test
independent dependent df =
independence variables are (right-tailed)
dependent.

- Observations are df= k-1-e discrete variable:


independent. k=number of cat.
To determine if e=number of
a sample estimators used. continuous variable:
- does not GOF-Test
GOF matches a follows
follows (right-tailed)
theoretical
Possible Estimators:
distribution.
Depending on context: first
; and last maybe or =

P(Type 1 error) = P(Reject | true) =


probability of X more extreme than c calculated with H0

P(Type 2 error) = P(Accept | true)


probability of X less extreme than c calculated with H1

Make a decision:
t more extreme than c OR p-value reject H0
H0 not rejected: we do not have enough evidence to reject H0

You might also like