What is a probability? Two schools in statistics: frequentists and Bayesians.

Similar documents
BAYESIAN HYPOTHESIS TESTING WITH SPSS AMOS

Bayesian and Frequentist Approaches

MS&E 226: Small Data

Introduction. Patrick Breheny. January 10. The meaning of probability The Bayesian approach Preview of MCMC methods

A Brief Introduction to Bayesian Statistics

Bayesian performance

Introduction to Bayesian Analysis 1

Bayesian Inference Bayes Laplace

A Case Study: Two-sample categorical data

Bayesians methods in system identification: equivalences, differences, and misunderstandings

Handout on Perfect Bayesian Equilibrium

Bayesian Estimation of a Meta-analysis model using Gibbs sampler

Ordinal Data Modeling

Att vara eller inte vara (en Bayesian)?... Sherlock-conundrum

Hierarchy of Statistical Goals

Bayes Theorem Application: Estimating Outcomes in Terms of Probability

MEASURING THE UNDIAGNOSED FRACTION:

ST440/550: Applied Bayesian Statistics. (10) Frequentist Properties of Bayesian Methods

A COMPARISON OF BAYESIAN MCMC AND MARGINAL MAXIMUM LIKELIHOOD METHODS IN ESTIMATING THE ITEM PARAMETERS FOR THE 2PL IRT MODEL

You must answer question 1.

Cognitive Modeling. Lecture 12: Bayesian Inference. Sharon Goldwater. School of Informatics University of Edinburgh

Introduction to Machine Learning. Katherine Heller Deep Learning Summer School 2018

CSE 258 Lecture 1.5. Web Mining and Recommender Systems. Supervised learning Regression

Bayesian Logistic Regression Modelling via Markov Chain Monte Carlo Algorithm

An Exercise in Bayesian Econometric Analysis Probit and Linear Probability Models

Model calibration and Bayesian methods for probabilistic projections

CAPL 2 Questionnaire

Introductory Statistical Inference with the Likelihood Function

This research is funded by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice (2011-WG-BX-0005).

UNLOCKING VALUE WITH DATA SCIENCE BAYES APPROACH: MAKING DATA WORK HARDER

Learning from data when all models are wrong

ERA: Architectures for Inference

Sensory specific satiation: using Bayesian networks to combine data from related studies

Journal of Clinical and Translational Research special issue on negative results /jctres S2.007

How to Choose the Wrong Model. Scott L. Zeger Department of Biostatistics Johns Hopkins Bloomberg School

Patrick Breheny. January 28

Probabilistic Modeling to Support and Facilitate Decision Making in Early Drug Development

Bayesian (Belief) Network Models,

Supplementary notes for lecture 8: Computational modeling of cognitive development

Hierarchical Bayesian Modeling of Individual Differences in Texture Discrimination

Practical Bayesian Design and Analysis for Drug and Device Clinical Trials

Combining Risks from Several Tumors Using Markov Chain Monte Carlo

Using historical data for Bayesian sample size determination

Lecture Outline Biost 517 Applied Biostatistics I. Statistical Goals of Studies Role of Statistical Inference

A Bayesian alternative to null hypothesis significance testing

Applications with Bayesian Approach

Biost 590: Statistical Consulting

Inference Methods for First Few Hundred Studies

The random variable must be a numeric measure resulting from the outcome of a random experiment.

Genome-Wide Localization of Protein-DNA Binding and Histone Modification by a Bayesian Change-Point Method with ChIP-seq Data

CISC453 Winter Probabilistic Reasoning Part B: AIMA3e Ch

Bayesian Confidence Intervals for Means and Variances of Lognormal and Bivariate Lognormal Distributions

P E R S P E C T I V E S

Lecture Outline Biost 517 Applied Biostatistics I

Advanced Bayesian Models for the Social Sciences

Law and Statistical Disorder: Statistical Hypothesis Test Procedures And the Criminal Trial Analogy

STATISTICAL INFERENCE 1 Richard A. Johnson Professor Emeritus Department of Statistics University of Wisconsin

Ecological Statistics

Bayes Factors for t tests and one way Analysis of Variance; in R

Vision as Bayesian inference: analysis by synthesis?

Professor Deborah G. Mayo

Applications of Bayesian methods in health technology assessment

From data to models: incorporating uncertainty into decision support systems. Outline. Probabilistic vs Mechanistic models.

How to Work with the Patterns That Sustain Depression

Bayesian vs Frequentist

Approximate Inference in Bayes Nets Sampling based methods. Mausam (Based on slides by Jack Breese and Daphne Koller)

Lisa A. Lundy, Page 1 of 5

Address : S. 27th Street, Suite 201 Franklin, Wisconsin, WebSite :

Fundamental Clinical Trial Design

Response to Comment on Cognitive Science in the field: Does exercising core mathematical concepts improve school readiness?

Dimensionality of the Force Concept Inventory: Comparing Bayesian Item Response Models. Xiaowen Liu Eric Loken University of Connecticut

Comparison of Meta-Analytic Results of Indirect, Direct, and Combined Comparisons of Drugs for Chronic Insomnia in Adults: A Case Study

Draft Methods Report Number XX

Pharmaceutical Statistics Journal Club 15 th October Missing data sensitivity analysis for recurrent event data using controlled imputation

The Century of Bayes

Beyond Subjective and Objective in Statistics

Institutional Ranking. VHA Study

Signal Detection Theory and Bayesian Modeling

Visit Names

Artificial Intelligence Programming Probability

Commentary: Practical Advantages of Bayesian Analysis of Epidemiologic Data

Practical and ethical advantages of Bayesian approaches in adaptive clinical trial designs. Kristian Thorlund

WORKSHOP 1 / ATELIER 1

Modelling crime linkage with Bayesian Networks

Bayesian Statistics Estimation of a Single Mean and Variance MCMC Diagnostics and Missing Data

Bayesian Joint Modelling of Benefit and Risk in Drug Development

Type and quantity of data needed for an early estimate of transmissibility when an infectious disease emerges

The Human Side of Science: I ll Take That Bet! Balancing Risk and Benefit. Uncertainty, Risk and Probability: Fundamental Definitions and Concepts

Ross Jeffries Speed Seduction

Statistical Hocus Pocus? Assessing the Accuracy of a Diagnostic Screening Test When You Don t Even Know Who Has the Disease

Increasing plasma donation frequency: Insights from current donors

A Comparison of Methods of Estimating Subscale Scores for Mixed-Format Tests

Bayesian Hierarchical Models for Fitting Dose-Response Relationships

Reliability and Validity

I. INTRODUCING CROSSCULTURAL RESEARCH

The Expanding Value of Biomarkers in NSCLC Treatment

Bayesian Phylogenetics Nick Matzke

Bayesian Adjustments for Misclassified Data. Lawrence Joseph

Chapter 13. Experiments and Observational Studies. Copyright 2012, 2008, 2005 Pearson Education, Inc.

Exploration of real-time crash likelihood of Powered-Two Wheelers in Greece. Road safety; real-time data; crash likelihood; powered-two-wheelers

Transcription:

Faculty of Life Sciences Frequentist and Bayesian statistics Claus Ekstrøm E-mail: ekstrom@life.ku.dk Outline 1 Frequentists and Bayesians What is a probability? Interpretation of results / inference 2 Comparisons 3 Markov chain Monte Carlo Slide 2 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics What is a probability? Two schools in statistics: frequentists and Bayesians. Slide 3 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics

Frequentist school School of Jerzy Neyman, Egon Pearson and Ronald Fischer. Slide 4 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Bayesian school School of Thomas Bayes P(H D)= P(D H) P(H) P(D H) P(H)dH Slide 5 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Frequentists Frequentists talk about probabilities in relation to experiments with a random component. Relative frequency of an event, A, is defined as P(A)= number of outcomes consistent with A number of experiments The probability of event A is the limiting relative frequency. Relative frequency 0.0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 80 100 n Slide 6 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics

Frequentists 2 The definition restricts the things we can add probabilities to: What is the probability of there being life on Mars 100 billion years ago? We assume that there is an unknown but fixed underlying parameter, θ, for a population (i.e., the mean height on Danish men). Random variation (environmental factors, measurement errors,...) means that each observation does not result in the true value. Slide 7 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics The meta-experiment idea The meta-experiment idea 167.2 cm

The meta-experiment idea 167.2 cm 175.5 cm The meta-experiment idea 167.2 cm 175.5 cm 187.7 cm The meta-experiment idea 167.2 cm 175.5 cm 187.7 cm 182.0 cm

Confidence intervals Thus a frequentist believes that a population mean is real, but unknown, and unknowable, and can only be estimated from the data. Knowing the distribution for the sample mean, he constructs a confidence interval, centeredatthesamplemean. Either the true mean is in the interval or it is not. Can t say there s a 95% probability (long-run fraction having this characteristic) that the true mean is in this interval, because it s either already in, or it s not. Reason: true mean is fixed value, which doesn t have a distribution. The sample mean does have a distribution! Thus must use statements like 95% of similar intervals would contain the true mean, if each interval were constructed from a different random sample like this one. Slide 9 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Maximum likelihood How will the frequentist estimate the parameter? Slide 10 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Maximum likelihood How will the frequentist estimate the parameter? Answer: maximum likelihood. Slide 10 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics

Maximum likelihood How will the frequentist estimate the parameter? Answer: maximum likelihood. Basic idea Our best estimate of the parameter(s) are the one(s) that make our observed data most likely. We know what we have observed so far (our data). Our best guess would therefore be to select parameters that make our observations most likely. Binomial distribution: P(Y = y)= ( ) n p y (1 p) n y y Slide 10 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Bayesians Each investigator is entitled to his/hers personal belief... the prior information. No fixed values for parameters but a distribution. Thumb tack pin pointing down: All distributions are subjective. Yours is as good as mine. Can still talk about the mean butitisthemeanofmy distribution. In many cases trying to circumvent by using vague priors. Prior distribution 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0.0 0.2 0.4 0.6 0.8 1.0 Theta Slide 11 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Credibility intervals Bayesians have an altogether different world-view. They say that only the data are real. The population mean is an abstraction, and as such some values are more believable than others based on the data and their prior beliefs. Slide 12 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics

Credibility intervals Bayesians have an altogether different world-view. They say that only the data are real. The population mean is an abstraction, and as such some values are more believable than others based on the data and their prior beliefs. The Bayesian constructs a credibility interval, centerednear the sample mean, but tempered by prior beliefs concerning the mean. Now the Bayesian can say what the frequentist cannot: There is a 95% probability (degree of believability) that this interval contains the mean. Slide 12 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Comparison Advantages Disadvantages Frequentist Objective Confidence intervals (not quite the desired) Calculations Bayesian Credibility intervals (usually the desired) Complex models Subjective Calculations Slide 13 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics In summary Afrequentistisapersonwhoselong-runambitionisto be wrong 5% of the time. A Bayesian is one who, vaguely expecting a horse, and catching a glimpse of a donkey, strongly believes he has seen a mule. Slide 14 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics

In summary Afrequentistisapersonwhoselong-runambitionisto be wrong 5% of the time. A Bayesian is one who, vaguely expecting a horse, and catching a glimpse of a donkey, strongly believes he has seen a mule. Afrequentistusesimpeccablelogictoanswerthe wrong question, while a Bayesean answers the right question by making assumptions that nobody can fully believe in. P. G. Hamer Slide 14 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Jury duty Slide 15 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics Example: speed of light What is the speed of light in vacuum really? Results (m/s) 299792459.2 299792460.0 299792456.3 299792458.1 299792459.5 Slide 16 PhD (Aug 23rd 2011) Frequentist and Bayesian statistics