9.3
9.4
9.5
9.6
9.7
tl
1 ) . 21 ( ' o t t t p t t l ; i o t t .
.
(
N l o < l c l) l r t x r k i r r g
\).2.1t
9.2.6 AnExarnple
M odel S electionU sing Z ellner'sg P rior
S urvival M odeling
F\rrtherReading
SummaryofRFunctions.
Exercises
'Jll7
.2cfw_)7
.'
STAT3120
Applied Bayesian Methods
Semester 2, 2013
Dr Frank Tuyl / Dr Darfiana Nur
School of Mathematical and Physical Sciences
The University of Newcastle, Australia
Welcome
2
Lecturers:
Dr Frank Tuyl (Course Coordinator, Weeks 7-12)
Frank.Tuyl@newcastl
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 8 Solutions
Q1 In this question we will illustrate that all cdf function
R Reference Card
by Tom Short, EPRI PEAC, tshort@epri-peac.com 2004-11-07
Granted to the public domain. See www.Rpad.org for the source and latest
version. Includes material from R for Beginners by Emmanuel Paradis (with
permission).
Getting help
Most R f
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 10 Solutions
Q1 Consider the case where we wish to simulate from f~N(0,1
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 9 Solutions
Q1
If you want WinBUGS for your home PC
WinBUGS was develope
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 9
Q1 Introduction to WINBUGS (Gibbs Sampler).
See Tutorial Introduction
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 11
Q1 Revisiting the binomial regression example in Lecture 8.
Log dosag
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 8
Q1
In this question we will illustrate that all cdf functions really a
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 10
Q1 Accept-Reject algorithm
Consider the case where we wish to simulat
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 7 Solutions
Q1 Consider the following 6 samples of counts of rejections
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 7
Q1 Consider the following 6 samples of counts of rejections after hear
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 6 Solutions
Q1 The table below is taken from BDA. It contains results of
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 Applied Bayesian Methods
Solutions to Lab 2
Q1 Albert, 1997, Chapter 3.4, Exercise 5: A geneticist is investiga
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 7
Q1 Consider the following 6 samples of counts of rejections after hear
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 3 Solutions
Q1 We looked at this question last week. Use Monte Carlo sim
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab 1 Introduction to Bayesian thinking
Introduction to R (Jim Albert, Bayesian Comput
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT 3120 APPLIED BAYESIAN METHODS
Lab 6
Q1 The table below is taken from BDA, chapter 3, exercise 6. It contains result
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab Exercises 5 Solutions
Q1
(a) The variance is known.
Suppose that we have a sample
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT 3120 APPLIED BAYESIAN METHODS
Lab 5
Q1 Suppose that we have a sample of size n, independently generated from a N(,2
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120-APPLIED BAYESIAN METHODS
Lab Exercises 4 Solutions
Q1
(i) and (ii)
1 ( xi ) 2
exp
f ( x1 , x 2 ,., x n | , )
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 - APPLIED BAYESIAN METHODS
Lab 4
Q1 [Classical method revision]
If x1, x2, xn are a random sample from a popula
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 Applied Bayesian Methods
Solutions to Lab 1
Open up R on your computer.
Introduction to R (Jim Albert, Bayesian
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab 3 - Monte Carlo inference and integration
Q1 We looked at this question last week.
FACULTY OF SCIENCE AND INFORMATION TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Lab 2
Q1 Albert, 1997, Chapter 3.4, Exercise 5: A geneticist is investigating the link
FACULTY OF SCIENCE AND INFORMATION
TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
LECTURE 11
11.1 Advanced MCMC with examples
In this lecture we will use some MCMC tech
FACULTY OF SCIENCE AND INFORMATION TECHNOLOGY
School of Mathematical and Physical Sciences
Student Enquiries
Telephone: (02) 4921-5000
STAT3120 APPLIED BAYESIAN METHODS
Semester 2, 2013
AIMS of LECTURE 10
Accept-reject algorithm (Lecture 9)
The Metropolis
LECTURE 8
Linear Regression
Linear regression is one of the most popular models utilised by quantitative data
analysts today.
Consider the ordinary linear regression model;
y X e ; e ~ N (0, 2 I )
where
o there are potentially q regressors in the model; q
STAT3120 APPLIED BAYESIAN METHODS
Semester 2, 2013
AIMS of LECTURE 11
Bioassay example (Lecture 11)
Combination MCMC and The M-H algorithm (We did this in Lecture 10)
The Metropolis algorithm (Lecture 11)
The Metropolis-Hastings (M-H) algorithm (Lecture