bayesian inference using jasp eric-jan wagenmakers

46
Bayesian Inference Using JASP Eric-Jan Wagenmakers

Upload: gwenda-powers

Post on 11-Jan-2016

238 views

Category:

Documents


5 download

TRANSCRIPT

Page 1: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayesian Inference Using JASP

Eric-Jan

Wagenmakers

Page 2: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 3: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Outline

Bayesian inference Bayesian parameter estimation Bayesian hypothesis testing The Bayesian t-test Example: Turning the hands of time

Page 4: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayesian Inferencein a Nutshell

In Bayesian inference, uncertainty or degree of belief is quantified by probability.

Prior beliefs are updated by means of the data to yield posterior beliefs.

Page 5: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Outline

Bayesian inference Bayesian parameter estimation Bayesian hypothesis testing The Bayesian t-test Example: Turning the hands of time

Page 6: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 7: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 8: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 9: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 10: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 11: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 12: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 13: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 14: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 15: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 16: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Outline

Bayesian inference Bayesian parameter estimation Bayesian hypothesis testing The Bayesian t-test Example: Turning the hands of time

Page 17: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayesian Hypothesis Test

Suppose we have two models, H0 and H1.

Which model is better supported by the data? The model that predicted the data best! The ratio of predictive performance is known

as the Bayes factor (Jeffreys, 1961).

Page 18: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayesian Hypothesis Tests

Page 19: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayesian Hypothesis Tests

Page 20: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayesian Hypothesis Tests

Page 21: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayesian Hypothesis Tests

Page 22: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayes Factor Inverse

Page 23: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayes Factor Transitivity

Page 24: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayes Factor Transitivity

Page 25: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Guidelines for Interpretation of the Bayes Factor

BF Evidence

1 – 3 Anecdotal 3 – 10 Moderate10 – 30 Strong 30 – 100 Very strong >100 Extreme

Page 26: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Visual Interpretation of the Bayes Factor

Page 27: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Visual Interpretation of the Bayes Factor

Page 28: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Visual Interpretation of the Bayes Factor

Page 29: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Advantages of the Bayes Factor

Quantifies evidence instead of forcing an all-or-none decision.

Allows evidence to be monitored as data accumulate.

Able to distinguish between “data support H0” and “data are not diagnostic”.

Page 30: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 31: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Disadvantages of the Bayes Factor

Where are the course books for psychologists?

Where is the software?

Page 32: Bayesian Inference Using JASP Eric-Jan Wagenmakers

August 6 & August 7, 2015University of Amsterdam

First Annual JASP WorkshopA Fresh Way to do Bayesian Statistics

jasp-stats.org

Page 33: Bayesian Inference Using JASP Eric-Jan Wagenmakers

August 10 - August 14, 2015University of Amsterdam

Fifth Annual JAGS and WinBUGS WorkshopBayesian Modeling for Cognitive Science

http://bayescourse.socsci.uva.nl/

Page 34: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 35: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 36: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Outline

Bayesian inference Bayesian parameter estimation Bayesian hypothesis testing The Bayesian t-test Example: Turning the hands of time

Page 37: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayes Factor for the t Test

Predictive Success for the Null Hypothesis

Predictive Success for the Alternative Hypothesis

Page 38: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Bayes Factor for the t Test

Prob. of Data Under the Null Hypothesis

Prob. of Data Under the Alternative Hypothesis

H0 states that effect size δ = 0. But how do we specify H1?

Page 39: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Effect size δ under H1

δ = .1

δ = .3

δ = .5

•••

•••

•••

Likelihood ratio

p(data | H0)

p(data | δ = .1)

p(data | H0)

p(data | δ = .3)

p(data | H0)

p(data | δ = .5)

The Bayes factoris the weighted average of the likelihood ratios.The weights are given by the prior plausibility assigned to the effect sizes.

Page 40: Bayesian Inference Using JASP Eric-Jan Wagenmakers

So we need to assign weight to the different values of effect size.

For complicated reasons, a popular default choice is to assume a Cauchy distribution (like a Normal, but with fatter tails):

These weigths reflect the relative plausibility of the effect sizes before seeing the data.

Page 41: Bayesian Inference Using JASP Eric-Jan Wagenmakers
Page 42: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Outline

Bayesian inference Bayesian parameter estimation Bayesian hypothesis testing The Bayesian t-test Example: Turning the hands of time

Page 43: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Turning the Wheels of Time

Topolinski and Sparenberg (2012): clockwise movements induce psychological states of temporal progression and an orientation toward the future and novelty.

Concretely: participants who turn kitchen rolls clockwise report more openness to experience.

Page 44: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Turning the Wheels of Time

Page 45: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Turning the Wheels of Time

Let's demonstrate:

– How to run a Bayesian t-test in JASP;

– How to interpret the output;

– How to conduct a sequential analysis;

– How to assess the robustness of the result.

Page 46: Bayesian Inference Using JASP Eric-Jan Wagenmakers

Conclusions

Bayesian hypothesis tests have a number of practical advantages;

These advantages are easily available through JASP;

The example discussed here was simple, but JASP handles more complicated analyses as well!