of 22

• date post

13-Apr-2017
• Category

## Education

• view

634

2

Embed Size (px)

### Transcript of Slide5 bme adat_2015

• session#5 Divnyi Jnos @divenyi.janos

• b.socrative.com

• How to find answersto relevant questions

using data

• QUESTION

DATA

ANALYSIS

PRESENTATION

• QUESTION

DATA

• QUESTION

DATA

ANALYSIS

• P(A|B)

conditional probability

probability that A occurs given that B has occurred

• E(Y|X)

cond. expected value

expected value of variable Ygiven that X takes a given value

• E(Y) = a + bX

linear regression

if E(Y|X) is linear

the regression function is it

• The Case of Berkeley (1973)

University of California, Berkeleywas sued for bias against women

• Men Women

ALL 2590 46% 1835 30%

ALL 2590 46% 1835 30%

• Men Women

ALL 2590 46% 1835 30%

• DepartmentMen Women

A

B

C

D

E

F

ALL 2590 46% 1835 30%

• DepartmentMen Women

A 825 62% 108 82%

B 560 63% 25 68%

C 325 37% 593 34%

D 417 33% 375 35%

E 191 28% 393 24%

F 272 6% 341 7%

ALL 2590 46% 1835 30%

• DepartmentMen Women

A 825 62% 108 82%

B 560 63% 25 68%

C 325 37% 593 34%

D 417 33% 375 35%

E 191 28% 393 24%

F 272 6% 341 7%

ALL 2590 46% 1835 30%

• xkcd.com/552

• Mr. Allen, the candidate for politicalParty A will run against Mr. Baker of Party B for office. Past races betweenthese parties for this office werealways close, and it seems that thisone will be no exception Party A candidates always have gottenbetween 40% and 60% of the vote and have won about half of the elections.

Carl Morris (1987)

• QUESTION

Mr. Allen needs to know whether he is going to win the election, so he ordersa poll. Which of three outcomes wouldbe the most encouraging for him?

(a) Y = 15, n = 20

(b) Y = 115, n = 200

(c) Y = 1046, n = 2000

Carl Morris (1987)