John Bardsley, University of Montana
Bayesian Scientic Computing
Accompanies Chapter 9 of the text
Bayesian Scientic Computing, Calvetti & Somersalo
Directed graph (internet search)
We consider the example in Fig. 9.1 from the text
Question: What node is
John Bardsley, University of Montana
Bayesian Scientic Computing
Accompanies Chapter 3 of the text
Bayesian Scientic Computing, Calvetti & Somersalo
Maximum Likelihood Estimation
Problem: You have data y from a random vector Y. The
data depends upon unkno
John Bardsley, University of Montana
Bayesian Scientic Computing
Accompanies Chapter 5 of the text
Bayesian Scientic Computing, Calvetti & Somersalo
Basic Problem of Statistical Inference
To infer on an unknown distribution based on a sample that
is belie
John Bardsley, University of Montana
Bayesian Scientic Computing
Accompanies Chapter 1 of the text
Bayesian Scientic Computing, Calvetti & Somersalo
Some Denitions
INVERSE PROBLEM: The problem of retrieving information
of unknown quantities by indirect me
John Bardsley, University of Montana
Bayesian Scientic Computing
Accompanies Chapter 2 of the text
Bayesian Scientic Computing, Calvetti & Somersalo
Fundamental Problem
Given a set of realizations from a random variable X
S = cfw_x1, x2, . . . , xN ,
xj R
John Bardsley, University of Montana
Bayesian Scientic Computing
Accompanies Chapter 10 of the text
Bayesian Scientic Computing, Calvetti & Somersalo
An Example
Suppose f : [0, 1] R is a continuous signal discretized as
follows:
j
xj = f (tj ), tj = , 0 j