June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng...
-
date post
19-Dec-2015 -
Category
Documents
-
view
213 -
download
0
Transcript of June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng...
![Page 1: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/1.jpg)
April 18, 2023 1
MARKOV CHAIN MONTE CARLO:A Workhorse for Modern Scientific Computation
Xiao-Li MengDepartment of Statistics
Harvard University
![Page 2: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/2.jpg)
2April 18, 2023
Introduction
![Page 3: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/3.jpg)
3April 18, 2023
Applications of Monte Carlo
Physics
Chemistry
Astronomy
Biology
Environment
Engineering
Traffic
Sociology
Education
Psychology
Arts
Linguistics
History
Medical Science
Economics
Finance
Management
Policy
Military
Government
Business
…
![Page 4: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/4.jpg)
4April 18, 2023
Monte Carlo 之应用A Chinese version of the previous slide
物理
化学
天文
生物
环境
工程
交通
社会
教育
心理
人文
语言
历史
医学
经济
金融
管理
政策
军事
政府
商务…
![Page 5: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/5.jpg)
5April 18, 2023
Monte Carlo Integration
Suppose we want to compute
where f(x) is a probability density. If
we have samples x1,…,xn » f(x), we can estimate I by
![Page 6: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/6.jpg)
6April 18, 2023
Monte Carlo Optimization
We want to maximize p(x) Simulate from
f(x) / p(x).
As ! 1, the simulated
draws will be more and
more concentrated around
the maximizer of p(x)
-5 0 5
0.0
0.1
0.2
0.3
0.4
x
dens
ity
-5 0 5
0.00
0.05
0.10
0.15
x
dens
ity
-5 0 50.0
e+00
8.0
e-09
x
dens
ity
=1
=2
=20
![Page 7: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/7.jpg)
7April 18, 2023
Simulating from a Distribution
What does it mean?Suppose a random variable (随机变量 ) X can only take two values:
Simulating from the distribution of X means that we want a collection of 0’s and 1’s:
such that about 25% of them are 0’s and about 75%of them are 1’s, when n, the simulation size is large.
The {xi, i = 1,…,n} don’t have to be independent
![Page 8: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/8.jpg)
8April 18, 2023
Simulating from a Complex Distribution
Continuous variable X, described by a density function f(x)
Complex: the form of f(x) the dimension of x
![Page 9: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/9.jpg)
9April 18, 2023
Markov Chain Monte Carlo
where {U(t), t=1,2,…} are identically and independently distributed.
Under regularity conditions,
So We can treat {x(t), t= N0, …, N} as an approximate sample from f(x), the stationary/limiting distribution.
![Page 10: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/10.jpg)
10April 18, 2023
Gibbs Sampler
Target density: We know how to simulate form the conditional
distributions
For the previous example,
N(,2)Normal Distribution
“Bell Curve”
![Page 11: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/11.jpg)
11April 18, 2023
![Page 12: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/12.jpg)
12April 18, 2023
Statistical Inference Point Estimator:
Variance Estimator:
Interval Estimator:
![Page 13: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/13.jpg)
13April 18, 2023
Gibbs Sampler (k steps)
Select an initial value (x1(0), x2
(0) ,…, xk(0)).
For t = 0,1,2, …, N Step 1: Draw x1
(t+1) from f(x1|x2(t), x3
(t),…, xk(t))
Step 2: Draw x2(t+1) from f(x2|x1
(t+1), x3(t),…, xk
(t))
…
Step K:Draw xk(t+1) from f(xk|x1
(t+1), x2(t+1),…, xk-1
(t+1))
Output {(x1(t), x2
(t),…, xk(t) ): t= 1,2,…,N}
Discard the first N0 draws
Use {(x1(t), x2
(t),…, xk(t) ): t= N0+1,2,…,N} as (approximate) samples from
f(x1, x2,…, xk).
![Page 14: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/14.jpg)
14April 18, 2023
Data Augmentation
We want to simulate from
But this is just the marginal distribution of
So once we have simulations:
{(x(t), y(t): t= 1,2,…,N)},
we also obtain draws:
{x(t): t= 1,2,…,N)}
0 2 4 6
0.0
0.1
0.2
0.3
0.4
0.5
0.6
x
de
nsi
ty
![Page 15: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/15.jpg)
15April 18, 2023
A More Complicated Example
![Page 16: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/16.jpg)
16April 18, 2023
Metropolis-Hastings algorithm
Simulate from an approximate distribution q(z1|z2), then
Step 0: Select z(0);
Now for t = 1,2,…,N, repeat
Step 1: draw z1 from q(z1|z2=z(t))
Step 2: Calculate
Step 3: set
Discard the first N0 draws
Accept
reject
![Page 17: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/17.jpg)
17April 18, 2023
M-H Algorithm: An Intuitive Explanation
Assume , then
![Page 18: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/18.jpg)
18April 18, 2023
M-H: A Terrible Implementation
We choose q(z|z2)=q(z)=(x-4)(y-4)
Step 1: draw x » N(4,1), y » N(4,1);
Dnote z1=(x,y)
Step 2: Calculate
Step 3: draw u » U[0,1]
Let
![Page 19: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/19.jpg)
19April 18, 2023
Why is it so bad?
![Page 20: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/20.jpg)
20April 18, 2023
M-H: A Better Implementation
Starting from some arbitrary (x(0),y(0))
Step 1: draw x » N(x(t),1), y » N(y(t),1)“random walk”
Step 2: dnote z1=(x,y), calculate
Step 3: draw u » U[0,1]Let
![Page 21: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/21.jpg)
21April 18, 2023
Much Improved!
![Page 22: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/22.jpg)
22April 18, 2023
Further Discussion
How large should N0 and N be?
Not an easy problem! Key difficulty:
multiple modes in unknown area
We would like to know all (major) modes, as well as their surrounding mass.
Not just the global mode
We need “automatic, Hill-climbing” algorithms.
) The Expectation/Maximization (EM) Algorithm, which can be viewed as a deterministic version of Gibbs Sampler.
![Page 23: June 2, 2015 1 MARKOV CHAIN MONTE CARLO: A Workhorse for Modern Scientific Computation Xiao-Li Meng Department of Statistics Harvard University.](https://reader038.fdocuments.us/reader038/viewer/2022110322/56649d275503460f949fde51/html5/thumbnails/23.jpg)
23April 18, 2023
Drive/Drink Safely,
Don’t become a Statistic;
Go to Graduate School,
Become a Statistician!