Beyes III

Question 1

bayesian inference and bayesian hypothesis testing difference

Answer 1

bayesian inference is based on parameter, bayesian hypothesis testing is based on the hypothesis

Question 2

marginal likelihood in hypothesis testing

Answer 2

bovenaan, gaat over H1

Question 3

likelihood in hypothesis testing

Answer 3

onderaan, gaat over H0

Question 4

hoe zie je of iets one sided of two sided is?

Answer 4

straight line = two sided!!!
als het opeens omhoog gaat = one sided!!!

Question 5

parsimony

Answer 5

the specific predictions, when still accurate, get rewarded more!

Question 6

how do you choose the prior distribution

Answer 6

baseren op knowledge of previous studies or keep informative

Question 7

what is the same about the prior distribution and the parameter

Answer 7

it needs to be in the same domain!

Question 8

which domains are there?

Answer 8

proportion: [0,1]
correlation: [-1,1]
difference in means: [-∞, ∞]

Question 9

wat meet je bij een bayesian correlation

Answer 9

rho (p)

Question 10

wat meet je bij een bayesian proportion

Answer 10

theta θ

Question 11

3 vormen van Ha

Answer 11

H1: p =/= 0
H+: p > 0
H-: p< 0

Question 12

dus als je een rho ziet, welk domein is dat dan

Answer 12

correlation: -1 tot 1

-> stretched beta distribution

Question 13

hoe kan je de median krijgen

Answer 13

median = top van grafiek.
meaning that there is a 50% that rho is equal to or lower than the observed correlation

Question 14

95% credible interval

Answer 14

95% probability that ρ is between … and …, under this model

Question 15

larger sample size leads to

Answer 15

more narrow likelihood, posterior distribution and credible interval

-> we can make a more specific prediction (with a higher BF), with the same certainty

Question 16

δ =

Answer 16

delta, cohens d.
staat voor differences between groups

Question 17

wat meet je bij een bayesian t test

Answer 17

the differences between groups, delta

Question 18

welk domein is een t test

Answer 18

difference in means: dus [-∞, ∞]!

Question 19

welke distribution gebruik je bij dit domein: [-∞, ∞]

Answer 19

Cauchy distribution, (= t distribution) with df = 1

(a uniform distribution would not work, because the domain is infinite)

Question 20

The Cauchy distribution is governed by a single shape parameter
that determines how wide it is

Answer 20

oke

Question 21

what is the “uninformative distribution” for the t test (difference in means)

Answer 21

width of 0,707:

50% of the values of delta are within -0,707 and 0,707

Question 22

bayes robustness check

Answer 22

= sensitivity analysis: what would have happened if we chose a different value for the prior width?

je ziet dan een grafiek met alle possible values voor de width. als die recht is, dan zat er niet zoveel verschil tussen de mogelijke waardes

Question 23

Bayes vs. frequentism

Answer 23

Frequentist Hypothesis
testing
1) Specify alpha x
2)Assumptions
3) Hypothesis
4) Test statistic + sampling x
distribution
5) P-value x
6) Conclusion

Bayesian Hypothesis testing
1) Assumptions
2) Hypothesis
3) Set prior distribution x
4) Compute likelihood x
5) Bayes factor x
6) Conclusion

Question 24

doel van Bayes

Answer 24

compare the predictive quality of two models/hypotheses

Question 25

doel van frequentism

Answer 25

minimize type 1 error (falsily rejecting H0)

Question 26

hoevaak kijken beiden technieken naar de data

Answer 26

Bayes: meerdere keren, updaten
Frequentism: maar 1 keer kijken (want meerdere keren = grotere kans op type 1 error)

Question 27

beide methoden kijken naar…

Answer 27

assumpties

Question 28

welke is meer compleet van de methoden

Answer 28

frequentism

Question 29

verschillen in sampling plans tussen de methoden

Answer 29

• bayes: can keep on monitoring sampling plan, update when needed and stop when preferred
• frequentism: need to specifiy beforehand and you cannot deviate from it

Question 30

– ‘evidence for HA’
– ‘probability that H0 is true’

Answer 30

incorrect!!!

Question 31

wat is de definitie van de p waarde

Answer 31

‘The probability of these data or more extreme if H0 is true’

Question 32

“Absence of evidence” vs “evidence of absence”

Answer 32

absence of evidence: je kan niets bewijzen
evidence of absence: het effect is er gewoon niet

Question 33

preregistration

Answer 33

Before you conduct your study, you preregister several pieces of information:
 Description of study, sampling plan
 Hypotheses (+ one-sided/two-sided)
 What statistical tests, and what settings: prior and Bayes factor / alpha-level for
accepting or rejecting hypothesis

Question 34

exploratory research

Answer 34

generate theories and hypotheses

Question 35

confirmatory research

Answer 35

test those theories and hypotheses from exploratory research

Question 36

what do journals do to promote preregistration

Answer 36

 Registered reports:
 Write most of your paper+preregistration, without conclusions, when it gets
accepted by the journal you conduct the experiment fill in the results and conclusion
section of your paper based on the results
 Increased support for null findings

Question 37

robustness check/sensitivity analysis

Answer 37

check the effect of the prior on the Bayes factor

Question 38

the BF is…

Answer 38

relative!!!

Question 39

Bayes does not fix all your problems: equally vulnerable to
assumption violations and statistical malpractice!

Answer 39

oke

Question 40

what 2 things can help with bettering science

Answer 40

• preregistration
• open science

Question 41

verschil sampling distribution and posterior distribution

Answer 41

• sampling: what would happen if we repeat over and over
• posterior: how well do all the values of the parameters predict the data? (prior + likelihood = posterior)

Question 42

verschil confidence interval and credible interval

Answer 42

• confidence interval: if we repeated this over and over, x% of the intervals would contain the true value
• credible interval: x% probability that this interval contains the true value