- My main page
- My talks
- My teaching notes (for resources not stored centrally)
- My statistics page
- My TensorFlow page

This web page lists my resources (notes / materials / examples) related to statistics.

- sPlots: This is a re-weighting technique used in particle physics, especially when validating multidimensional fit models. An example ROOT macro can be downloaded from here, along with this brief overview of what sPlots are and how to use them.

- Generalisation: This is the issue of trying to determine a robust MVA training that can be applied to an unseen data sample and be expected to perform well; as opposed to being fine tuned (over trained/over fitted) on statistical fluctuations within the training sample itself. Generalisation.pdf provides a brief introduction to the issue and discusses decay-weight regularisation for neural networks as well as cross validation algorithms.
- Support Vector Machines: SVMs.pdf provides a brief introduction to the soft and hard margin SVM algorithms and exemplifies their application using the example of a checker board in TMVA.
- R example scripts
- Neural network example. Using an MLP for regression: neuralNetwork.r
- SVM example: svm.r

- Tensor Flow examples: Please see my TensorFlow page for some examples; including basic operations, plotting MNIST data, computing Fisher discriminant coefficients; plotting Fractals; training a perceptron and training an MLP.

- AFit: This is a ROOT (and RooFit)-based framework for performing extended unbinned maximum likelihood fits to data. The latest development version of AFit is now incorporated within the ATLAS experiment svn repostiory, where a number of additional features have been added for uses on that experiment. The basic framework is still of general use.

- Statistical Data Analysis: This is a 3rd year undergraduate lecture course on statistics that I teach. The material covered in this course complements that found in my book on statistics.
- BaBar Analysis School Lectures on MVA methods: This is a short introductory course on the use of multivariate methods. There is an example provided that uses data from the BaBar experiment in order to help you learn you way around the use of several techniques. These lectures were delivered at the SLAC National Accelerator Laboratory in California in 2009 and 2011.
- UK HEP Young Experimentalists and Theorists Institute on Statistical methods: Tutorials: In January 2007 I gave some tutorials on fitting and the use of multivariate methods for the UK HEP community. The material for these tutorials provides a potentially useful starting point for people wanting to learn how to use some of the tools available to the HEP community.

- Statistical Data Analysis for the Physcical Sciences: This is my statistics book, aimed and undergraduate and graduate students wanting to understand statistics at a level that covers what you need for undergraduate laboratory based data analysis. The latter chapters of this book cover more indepth material which may be of interest to people using statistics later in their career (in research or indistry).

Back to the *home-page*

Maintained by Adrian Bevan. | Last update on June-2014 |