Bayesian performance

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

Introduction to Bayesian Analysis 1

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

MS&E 226: Small Data

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

Learning from data when all models are wrong

Practical Bayesian Design and Analysis for Drug and Device Clinical Trials

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

Coherence and calibration: comments on subjectivity and objectivity in Bayesian analysis (Comment on Articles by Berger and by Goldstein)

A Brief Introduction to Bayesian Statistics

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

INTRODUCTION TO BAYESIAN REASONING

BAYESIAN HYPOTHESIS TESTING WITH SPSS AMOS

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

Discussion. Risto Lehtonen Introduction

Bayesian Inference. Review. Breast Cancer Screening. Breast Cancer Screening. Breast Cancer Screening

Meta-analysis of few small studies in small populations and rare diseases

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

Bayesian Inference Bayes Laplace

Bayesian Methods LABORATORY. Lesson 1: Jan Software: R. Bayesian Methods p.1/20

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

Using Item Response Theory To Rate (Not Only) Programmers

Bayes Theorem Application: Estimating Outcomes in Terms of Probability

Bayesian Adjustments for Misclassified Data. Lawrence Joseph

Response to the ASA s statement on p-values: context, process, and purpose

Bayesian and Frequentist Approaches

Bayesian Inference. Thomas Nichols. With thanks Lee Harrison

Bayesian Adjustments for Misclassified Data. Lawrence Joseph

Hierarchical Bayesian Modeling of Individual Differences in Texture Discrimination

What is probability. A way of quantifying uncertainty. Mathematical theory originally developed to model outcomes in games of chance.

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

Calibrated Bayes: A Bayes/Frequentist Roadmap

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

Understanding Statistics for Research Staff!

A Case Study: Two-sample categorical data

Doing Thousands of Hypothesis Tests at the Same Time. Bradley Efron Stanford University

How to use the Lafayette ESS Report to obtain a probability of deception or truth-telling

UW Biostatistics Working Paper Series

Missing data. Patrick Breheny. April 23. Introduction Missing response data Missing covariate data

Intelligent Systems. Discriminative Learning. Parts marked by * are optional. WS2013/2014 Carsten Rother, Dmitrij Schlesinger

Meta-analysis of two studies in the presence of heterogeneity with applications in rare diseases

"PRINCIPLES OF PHYLOGENETICS: ECOLOGY AND EVOLUTION"

3. Fixed-sample Clinical Trial Design

Part 1: Modelling and Estimation. Maximum Likelihood Estimation. A nonparametric regression smoother. Social Science and Parametric Models

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

An Exercise in Bayesian Econometric Analysis Probit and Linear Probability Models

Excursion 1: How to Tell What s True about Statistical Inference

Sections 10.7 and 10.9

Dynamic Causal Modeling

Natural Scene Statistics and Perception. W.S. Geisler

Reasoning with Uncertainty. Reasoning with Uncertainty. Bayes Rule. Often, we want to reason from observable information to unobservable information

15.301/310, Managerial Psychology Prof. Dan Ariely Recitation 8: T test and ANOVA

Undesirable Optimality Results in Multiple Testing? Charles Lewis Dorothy T. Thayer

Introduction to screening tests. Tim Hanson Department of Statistics University of South Carolina April, 2011

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

Statistical Decision Theory and Bayesian Analysis

Hierarchy of Statistical Goals

Model calibration and Bayesian methods for probabilistic projections

An Introduction to Bayesian Statistics

Using historical data for Bayesian sample size determination

BAYESIAN ESTIMATORS OF THE LOCATION PARAMETER OF THE NORMAL DISTRIBUTION WITH UNKNOWN VARIANCE

arxiv: v1 [math.st] 8 Jun 2012

Computer Age Statistical Inference. Algorithms, Evidence, and Data Science. BRADLEY EFRON Stanford University, California

INHERENT DIFFICULTIES WITH ACTIVE CONTROL EQUIVALENCE STUDIES*

Midterm project due next Wednesday at 2 PM

Sample Size Considerations. Todd Alonzo, PhD

T-Statistic-based Up&Down Design for Dose-Finding Competes Favorably with Bayesian 4-parameter Logistic Design

Bayesian Model Averaging for Propensity Score Analysis

Introductory Statistical Inference with the Likelihood Function

Cognitive Science and Bayes. The main question. Several views. Introduction. The main question Several views Examples. The heuristics view

Sampling, Modeling and Measurement Error in Inference from Clinical Text

Bayes theorem describes the probability of an event, based on prior knowledge of conditions that might be related to the event.

The Wellbeing Course. Resource: Mental Skills. The Wellbeing Course was written by Professor Nick Titov and Dr Blake Dear

Intro to Probability Instructor: Alexandre Bouchard

For general queries, contact

The Frequentist Implications of Optional Stopping on Bayesian Hypothesis Tests

Bayesian Analysis by Simulation

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

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

Citation for published version (APA): Ebbes, P. (2004). Latent instrumental variables: a new approach to solve for endogeneity s.n.

Standing Between a Bayesian and a Frequentist: An Emperical Bayes Exploration of Movies, Baseball, and Williams College.

Bayesian vs Frequentist

Signal Detection Theory and Bayesian Modeling

Acknowledgements. Section 1: The Science of Clinical Investigation. Scott Zeger. Marie Diener-West. ICTR Leadership / Team

A Bayesian Account of Reconstructive Memory

Agenetic disorder serious, perhaps fatal without

Beyond Subjective and Objective in Statistics

The Bayesian flip Correcting the prosecutor s fallacy

ROLE OF RANDOMIZATION IN BAYESIAN ANALYSIS AN EXPOSITORY OVERVIEW by Jayanta K. Ghosh Purdue University and I.S.I. Technical Report #05-04

I. Introduction and Data Collection B. Sampling. 1. Bias. In this section Bias Random Sampling Sampling Error

PHILOSOPHY OF STATISTICS: AN INTRODUCTION

Illustrating Frequentist and Bayesian Statistics in Oceanography 1

Macroeconometric Analysis. Chapter 1. Introduction

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

Section 1: The Science of Clinical Investigation

Representativeness heuristics

Bayesian Models for Combining Data Across Subjects and Studies in Predictive fmri Data Analysis

On Test Scores (Part 2) How to Properly Use Test Scores in Secondary Analyses. Structural Equation Modeling Lecture #12 April 29, 2015

Bayesian Models for Combining Data Across Domains and Domain Types in Predictive fmri Data Analysis (Thesis Proposal)

Transcription:

Bayesian performance In this section we will study the statistical properties of Bayesian estimates. Major topics include: The likelihood principle Decision theory/bayes rules Shrinkage estimators Frequentist properties of Bayesian estimators ST740 (2) Bayes Performance - Part 1 Page 1

The likelihood principle In a Bayesian analysis, everything we know about the parameters is summarized by the posterior (likelihood x prior). Is it true that classical methods only use the likelihood? Likelihood principle: Once the data are observed, the likelihood contains all the information in the data about the parameters. The p-value can violate the likelihood principle because it depends on both the data and unobserved events. Example (Lindley and Phillips): A coin with P(heads)=θ is flipped 12 times and we observe 9 heads. Now test H 0 : θ = 0.5 versus H 1 : θ > 0.5. Analysis 1: Analysis 2: ST740 (2) Bayes Performance - Part 1 Page 2

The likelihood principle A Bayesian analysis adheres to the likelihood principle. For example, in both analysis Do you think the likelihood principle is important? ST740 (2) Bayes Performance - Part 1 Page 3

Calibrated Bayes Now we ll begin to study the frequentist properties of Bayesian estimators (unbiasedness, consistency, etc.). First, should we (Bayesians) really care about frequentist properties of estimators? ST740 (2) Bayes Performance - Part 1 Page 4

Calibrated Bayes Some quotes from Little (2011). Calibrated Bayes, for Statistics in General, and Missing Data in Particular. Statistical Science, 26, 162-174. Little: To summarize, Bayesian statistics is strong for inference under an assumed model, but relatively weak for the development and assessment of models. Frequentist statistics provides useful tools for model development and assessment, but has weaknesses for inference under an assumed model. If this summary is accepted, then the natural compromise is to use frequentist methods for model development and assessment, and Bayesian methods for inference under a model. This capitalizes on the strengths of both paradigms, and is the essence of the approach known as Calibrated Bayes. Rubin: The applied statistician should be Bayesian in principle and calibrated to the real world in practice - appropriate frequency calculations help to define such a tie...frequency calculations are useful for making Bayesian statements scientific, scientific in the sense of capable of being shown wrong by empirical test; here the technique is the calibration of Bayesian probabilities to the frequencies of actual events. ST740 (2) Bayes Performance - Part 1 Page 5

In a Bayesian analysis all inference is based on the posterior distribution p(θ y). What is the best one-number summary of the posterior, ˆθ, to be used as the estimator? This depends on the situation, and in particular, on the penalty associated with different types of errors (e.g., maybe overestimation is way worse than underestimation). We will use decision theory to form estimators with good properties. ST740 (2) Bayes Performance - Part 1 Page 6

We need a definition of best to get started. Let θ 0 be the true value of the parameter and ˆθ(y) be our estimator (perhaps the posterior mean). The loss function l[θ 0, ˆθ(y)] is cost associated with estimating θ to be ˆθ(y) when the truth is θ 0. Examples: ST740 (2) Bayes Performance - Part 1 Page 7

The loss function l[θ 0, ˆθ(y)] depends on both the true value (θ 0 ) and the data (via ˆθ). We need to average over one or both of these to compare methods. Bayesian analysis is conditioned on the data, so we average over θ 0. Risk = l[θ, ˆθ(y)]w(θ)dθ. Which values of θ 0 should be weighted the highest? Bayesian risk = The Bayes rule is the estimator ˆθ(y) that minimizes Bayesian risk. ST740 (2) Bayes Performance - Part 1 Page 8

Under squared error loss l[θ 0, ˆθ(y)] = [θ 0 ˆθ(y)] 2, the Bayes rule is ST740 (2) Bayes Performance - Part 1 Page 9

Under absolute loss l[θ 0, ˆθ(y)] = θ 0 ˆθ(y), the Bayes rule is Under zero/one loss l[θ 0, ˆθ(y)] = I[θ 0 = ˆθ(y)], the Bayes rule is Hypothesis testing: Say θ = 0 if H 0 is true and θ = 1 if H 1 is true. Give the Bayes rule under the loss l[θ 0, ˆθ(y)] = λ 1 I[θ 0 = 0, ˆθ(y) = 1] + λ 2 I[θ 0 = 1, ˆθ(y) = 0]. How to pick λ 1 and λ 2? ST740 (2) Bayes Performance - Part 1 Page 10

We ve seen loss function for point estimation and hypothesis testing, which loss functions are appropriate for interval estimation? ST740 (2) Bayes Performance - Part 1 Page 11