Scribd
Upload
34 views10 pages

Marginal and Conditional Distributions

This document discusses marginal and conditional distributions and frequencies using a 2x2 contingency table. It defines marginal frequencies as the totals calculated by adding over subdivisions of an indexed class. For example, the marginal frequency of smokers is the total of smokers with lung cancer and smokers without lung cancer. Conditional frequencies show how the distribution of one variable changes given levels of another variable. In a 2x2 table with rows and columns, the conditional frequencies are the probabilities in each cell given the row or column. The document also generalizes these concepts from frequencies to probability distributions.

Uploaded by

antialonso
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
34 views10 pages

Marginal and Conditional Distributions

This document discusses marginal and conditional distributions and frequencies using a 2x2 contingency table. It defines marginal frequencies as the totals calculated by adding over subdivisions of an indexed class. For example, the marginal frequency of smokers is the total of smokers with lung cancer and smokers without lung cancer. Conditional frequencies show how the distribution of one variable changes given levels of another variable. In a 2x2 table with rows and columns, the conditional frequencies are the probabilities in each cell given the row or column. The document also generalizes these concepts from frequencies to probability distributions.

Uploaded by

antialonso
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 10

Marginal and Conditional

Distributions/frequencies
A brief accordance with the 2x2 contingency table

Presenter
Rishav Mukherjee
MSc. Biostatistics
Why marginal ?
Horizontal margins

LC No LC Total

Smoker a b a+b

Non c d c+d
Smoker
Total a+c b+d a+b+c
+d

Vertical margins
Marginal Frequencies for a 2x2 table

LC No LC Total
Marginal
Smoker a b a+b
frequency of
smokers
Non c d c+d Marginal
Smoker frequency of
non smokers
Total a+c b+d a+b+c
+d

Marginal frequency Marginal Frequency of


of people with lung people with no lung
cancer cancer
Note
The marginal frequencies are calculated by taking an index class and
adding over all the different sub divisions of that indexed class.
(In case of marginal frequency of smokers. Smokers is the indexed class
and the frequency is calculated by summing over the two subdivisions,
smokers who have lung cancer and smokers who do not have lung
cancer)
From frequencies to distributions
(The generalization to probability distributions)
Type of Random Discreet Continuous
Variable
Distribution

Joint Pr (x,y) = f(x,y)=

Marginal of X Pr(x) = = p0j fX (x) =


(assuming y runs from j=1(1)m)

Marginal of Y Pr(y) = = pi0 fY (y)=


(assuming x runs from i=1(1)n)
Why conditional ?
To calculate the frequency or density of a particular random variable
we are given a condition i.e the level of the other random variable(S)
is fixed to a given value or range.
P(x|y=20) = we have to find the probability of x, when the following
condition is given
Probability of y being 20
Conditional frequencies in 2x2 table
(Generally) in contingency tables where one variable (say Y)is a
response variable and the other (say X) is an explanatory variable
Used to study how the distribution or frequency of Y changes, with
changes in the different levels of X.
In a 2x2 table X has 2 levels and Y has 2 levels.
contd
Column 1 2 Total
Row

1 p11 p12 p10


(p1|1) (p2|1)

2 p21 p12 p20


(p1|2) (p2|2)

Total p01 p02


From frequencies to distributions
(The case of probability distributions)
Type of Random Discreet Continuous
Variable
Distribution

Joint Pr (x,y) = f(x,y)=

Conditional of X given Y Pr (x|y) = f(x|y)=

Conditional of Y given X Pr (y|x) = f(y|x)=


Thank You

You might also like