myESL

My slides and MATLAB codes in 2014 summer reading: the Elements of Statistical Learning. This repository is migrated from my Purdue GitHub account. And I will keep updating my practice notes on ESL here.

Chapter 8: Model Inference and Averaging

8.5 The EM Algorithm

I implemented the Algorithm 8.1, EM Algorithm for Two-component Gaussian Mixture, in Matlab/R2012b. Simply execute /Chap8/mcodes/EM.m, and you will see two figures:

• Figure 1, the histogram of data of the mixture example (see also Figure 8.5 in ESL book);
• Figure 2, EM learning curve (see also Figure 8.6 in ESL book).

8.6 MCMC for Sampling from the Posterior

Execute /Chap8/mcodes/Gibbs.m, and you will see the histogram as shown in the EM section, and the figure as shown in Figure 8.8 in ESL book.

Exercise 8.6 Bone mineral density data

Find the code /Chap8/mcodes/Ex86.m, and it fits the data using (a) cubic smooth spline, (b) Bayesian methods, and (c) bootstrap.

Chapter 17: Undirected Graphical Models

17.3.1 Modified Regression for UGM with Known Structure

See /Chap17/mcodes/Call171.m for my implementation of Algorithm 17.1.

17.3.2 Graphical Lasso

See /Chap17/mcodes/Call172.m for my implementation of Algorithm 17.2.