The module will cover key mathematical foundational concepts and methods relevant to Data Science, including: Sampling theory (Nyquist-Shannon)Non-square matrices; singular value decomposition (SVD); matrix norms; dualityMultivariate statistics: Introduction, Bayes theorem, priors, posteriors; hypothesis testing; frequentist versus BayesianCramer Rao criterion; distribution of p values; sampling distributions; non-parametric testing (e.g. odds ratio); advanced statistical methodsImplications of the Central Limit theorem Sampling algorithms (e.
g. Metropolis Hastings)
Learning Outcomes
By the end of the module students should be able to:
Apply mathematical descriptions of the information content of data, including Nyquist-Shannon sampling theory
Calculate small SVDs by hand and understand rank deficiency
Apply Bayes theorem in different contexts, e.g.
, to multivariate parameter estimation problems
Derive and apply suitable statistical tests
Interpret the implications of the central limit theorem
Be able to implement a sampling algorithm, and be cognisant of different techniques
Assessment
Assessment Methods & Exceptions
Assessment:
One 1.5 hour exam (80%) and continuous assessment (20%) [NOTE: under EPS exception to CoP]