Institute of Astronomy


Lent Term 2014

Sergey Koposov -  Statistical Techniques in Python 

Lectures 1-2:
    Introduction into Python
    Python datatypes
    Useful python packages and tools

Lectures 2-8:
    Statistics introduction.
    Understanding probabilities.
    Introduction of the Bayesian framework
    Maximum likelihood as an approximation of the full Bayesian analysis
    Measuring errors from the Maximum Likelihood fits
    Practical implementation of the ML fitting in Python
    Different ML optimizers in Python
    Fitting data with outliers
    Markov Chain Monte-Carlo methods
    Python packages for performing MCMC analysis
    Mixture models
    Gaussian Mixtures
    Model selection (AIC, cross-validation, Bayes factors)
    Classification using mixture models
    Simple machine learning classification algorithms in Python.
    Hierchical Bayesian models
    Building statistical models using STAN package.

Page last updated: 15 January 2015 at 11:40