Scribd
Upload
636 views4 pages

Răspuns: Binomial (N 150, T) Distribution

The document describes four examples of using binomial and beta distributions to model polling data and determine posterior distributions based on sample data and prior assumptions. In the first example, the posterior distribution for the proportion of people who saw a magazine issue is beta(30,122) based on a sample of n=150 people and y=29 seeing the issue. In the third example, the prior is set as beta(5.05,5.05) to match a mean of 0.5 and standard deviation of 0.15. The equivalent sample size is 11.11.

Uploaded by

Delia Radu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
636 views4 pages

Răspuns: Binomial (N 150, T) Distribution

The document describes four examples of using binomial and beta distributions to model polling data and determine posterior distributions based on sample data and prior assumptions. In the first example, the posterior distribution for the proportion of people who saw a magazine issue is beta(30,122) based on a sample of n=150 people and y=29 seeing the issue. In the third example, the prior is set as beta(5.05,5.05) to match a mean of 0.5 and standard deviation of 0.15. The equivalent sample size is 11.11.

Uploaded by

Delia Radu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

8.

1 In order to determine how effective a magazine is at reaching its target audience, a


market research company selects a random sample of people from the target audience and
interviews them. Out of the 150 people in the sample, 29 had seen the latest issue.
n=150
y=29
(a) What is the distribution of y, the number who have seen the latest issue?
Distribuția condiționată a observației Y, numărul total de succese în n încercări date cu
parametrul T, este binomial (n, T).

Răspuns: binomial (n = 150, T) distribution.

(b) Use a uniform prior for T, the proportion of the target audience that has seen the
latest issue. What is the posterior distribution of T?
Când examinăm acea parte a formulei care arată forma posterioară ca funcție a lui T,
recunoaștem că aceasta este o distribuție beta (a, b), unde:
a = y + 1 și b=n – y + 1. Așadar, avem: a = 29+1=30 și b=150-29+1=122.
Deci, în acest caz, se obține cu ușurință distribuția posterioară. Tot ce este necesar este să
ne uităm la exponenții lui T și (1 - T). Nu a trebuit să facem integrarea.
Deci, răspunsul corect este beta (30,122).

8.2 A city is considering building a new museum. The local paper wishes to determine the
level of support for this project, and is going to conduct a poll of city residents. Out of the
sample of 120 people, 74 support the city building the museum.
(a) What is the distribution of y, the number who support the building the
(b) Use a uniform prior for T, the proportion of the target audience that museum support
the museum. What is the posterior distribution of T?

(a) What is the distribution of y, the number who support the building the

(b) Use a uniform prior for π , the proportion of the target audience that support the museum. What is
the posterior distribution of π ?
A ) n=120

y=74

Distributie binomial (n=120, π ¿

f ( y|π ) = n π y ¿
() y
Raspuns :

f ( 74|π )= 120 π 74 ¿
( )
74

B ) n=120

y=74

Uniform prior for : beta(a,b) => beta(75,47).

a=y+1, b=n-y+1.

8.3 Sophie, the editor of the student newspaper, is going to conduct a survey of students to
determine the level of support for the current president of the students association. She
needs to determine her prior distribution for T, the proportion of students who support the
president. She decides her prior mean is .5, and her prior standard deviation is .15.
(a) Determine the beta(a, b) prior that matches her prior belief.
a
Prior mean este: m=
a+b
a
a și b sunt soluții simultane pentru m= =0,5
a+b
Avem beta(a, b) posterior, așadar, varianța posterioară este:
a×b
Var(π / y)= 2 =0,152
(a+ b) ×(a+ b+1)
a
=.5 => a=.5*(a+b) => a- .5a=.5b => 0,5a=0,5b => a=b
a+b
a×b
=.15 2 a2 a2 a2
2 => =0.0225 => =0,0225 => =0,0225
(a+ b) ×(a+ b+1) ( 2a )2 × (2 a+ 1 ) 2
4 a × ( 2a+ 1 ) 8 a3 ×4 a2
1
=> 8a+4= =>a=5,05 si b=5,05
0,0225
(b) What is the equivalent sample size of her prior?
Mărimea echivalentă a eșantionului este a + b + 1 = 5,05+5,05+1=11,11.

(c) Out of the 68 students that she polls, y = 21 support the current president.
Determine her posterior distribution.
Posterior distribution:

n=68
y=21
beta(a’,b’)
a’ = y + a și b’=n – y + a.
Așadar, avem: a’ = 21+5,05=26,05 și b’=68-21+5,05=52,05
Soluția este: beta (26.05,52.05)

8.4 You are going to take a random sample of voters in a city in order to estimate the
proportion T who support stopping the fluoridation of the municipal water supply. Before
you analyze the data, you need a prior distribution for T. You decide that your prior mean
is .4, and your prior standard deviation is .l .
(a) Determine the betu(a, b) prior that matches your prior belief.
(b) What is the equivalent sample size of your prior?
(c) Out of the 100 city voters polled, y = 21 support the removal of fluoridation from the
municipal water supply. Determine your posterior distribution.
Rezolvare :

a) Prior mean - .4

Std.dev.-.1

a 3a
=.4 => a=.4*(a+b) => a- .4a=.4b => b=
a+b 2
3 a2
a×b 2
2
=.12 => 2 =0.01 => a=9.2 si b=13.8
(a+ b) ×(a+ b+1) 5a 5 a+ 2
2 ( ) (
×
2 )
b) Sample size :

n eq=a+b+1=9.2+13.8+1=24

c) Posterior distribution : n=100, y =21

beta ( a' ,b ' )

Unde :

a ' = y+ a = 30.2

b ' =b+n− y =92.8


Rezulta : beta(30.2, 92.8 )

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy