Applied Statistics - Week 6

Monday the 2nd - Friday the 6th of January 2017

The following is a description of what we will go through during this week of the course. The chapter references and computer exercises are considered read, understood, and solved by the beginning of the following class, where I'll shortly go through the exercise solution.

General notes, links, and comments:

:

Monday:
We will have a short exercise about planning of experiments, using "A Lady tasting tea" as our source of inspiration, and then expanding from there. Part of class will also be dedicated to continue working on Project2.

Reading:

Lady tasting tea (Wikipedia).

Short note on Lady tasting tea.

NOTE: You should by now have read curriculum (roughly Barlow chapters 1-8).
Lecture(s):

There are no more formal lectures.
Computer Exercise(s):

Weldon's Dices: WeldonsDices.py.

Planning an Experiment: PlanExp.py


Tuesday:
We will use two days for the last "major" theme in these course, which is MultiVariate Analysis (MVA), that is analysis of data with more than one (typically many) variables. To begin with, we will consider the relatively simple linear case, which is described by Fisher's Discriminant, but Friday move on to more complex sets of data, for which more advanced non-linear methods, such as Neural Networks (NN) and Boosted Decision Trees (BDT) are more useful.

Reading:

Lecture(s):

AS2016_0103_MultiVariateAnalysis1.pdf
Computer Exercise(s):

2par_discriminant.py

fisher_discriminant.py and data

Friday:
Reading:

Lecture(s):

AS2016_0106_MultiVariateAnalysis2.pdf
Computer Exercise(s):

MVA_train_discriminant.py and
three TMVA files: TMVAGui.C, TMVAlogon.C, and tmvaglob.C

Data samples: DataSample.root, atlas_test_beam_data.root Higgs14TeV.root, ZZ14TeV.root

"Real life" TMVA setup example (with two simple variables): GenerateData.py, TrainClassifier.py, and UseClassifier.py.

Finally, a link to an online course on Machine Learning (by Udacity), which I was recommended.

---

Last updated: 7th of January 2017.