Computational Statistics, STP 540, Fall 2019

-log L


Course Information

Instructor/TA:

Instructor: Robert McCulloch, robert.mcculloch@asu.edu
office: Wexler 528
TA: 494: Xuetao Lu, xuetaolu@asu.edu

Time and Place:

STP 540, Computational Statistics
T Th 12:00 PM 1:15 PM Tempe - WXLR A103 08/22 - 12/06


Syllabus

Syllabus

Office hours: Thursday, 4:30-5:30, Wexler 528."


Suggested Projects

final due date: 9am, December 12.

Mixture Modeling with EM and MCMC

Inference for the parameters of a Gaussian Process

Monte Carlo EM


Homework

Here are instructions for how to submit homework using canvas: how to submit homework

Make sure you include each group members name clearly and whether they are registered for 494 or 598.

Homework 1, due September 17: Homework 1

Homework 2, due October 8: Homework 2
Robs rough code for problem 4


Homework 3, due November 7: Homework 3



Notes

The Multivariate Normal and the Choleski and Eigen Decompositions
   Look at cholesky and spectral in R


The EM Algorithm See Chapter 4 of Givens and Hoeting.


Monte Carlo See Chapter 6 of Givens and Hoeting.
   Geweke paper
   R script to try various truncated normal draws
   R script to try various importance sampling approaches for prior sensitivity
   Prior based on odds ratio
   SIR R script


Introduction to Bayesian Statistics:
   Introduction to Bayesian Statistics and the Beta/Bernoulli Inference
   Normal Mean Given Standard Deviation
   Normal Standard Deviation Given Mean
   Introduction to Bayesian Regression


MCMC: Markov Chain Monte Carlo,   See Chapter 7 of Givens and Hoeting.
   Markov Chains
     Simple Example of a Markov Chain
   Gibbs Sampling
     Note: Hoff refers to ``A First Course in Bayesian Statistical Methods'', by Peter Hoff
   Reversable Markov Chains
   The Metropolis Algorithm


State Space Models and FFBS  
   Hotels Problem
   intro to state space models
   FFBS
   R code for hotels example
   Forward Filtering for Simple Hotels model


The Bootstrap (Efron and Hastie, chapters 10 and 11)
   The Bootstrap

A Little Optimization (Givens and Hoeting 2.2.2)
   A Littel Optimization


R and Python

Information on R

Information on Python