20 credits

This introductory module is aimed at graduates who have not previously studied statistics at university level. It assumes no prior knowledge of statistics and builds from simple concepts to theoretical methods that are required for application to real life data and problems. The module will provide the foundation elements of probability and statistics that are required for the more advanced modules studied later on.

This will include:

• an introduction to probability and probability rules
• random variables and probability distributions
• data visualisation and representation
• hypothesis testing and confidence intervals
• power and sample size calculations
• correlation and simple linear regression

20 credits

This module will introduce the R computing environment and enable you to import data and perform statistical tests. The module will then focus on the understanding of the least squares multiple regression model, general linear model, transformations and variable selection procedures.

You can expect to cover concepts such as:

• use of functions and packages in R
• use of the tidyverse for data manipulation
• data visualisation using both base R and ggplot2
• multiple linear regression
• using variable selection techniques to cope with large data sets
• more general model comparison

10 credits

This module provides students with the fundamental principles of statistical modelling through experimental design. The statistical models used in the analysis of balanced experimental designs are derived and used in the analysis of data sets.

You will cover topics such as:

• completely randomised design
• randomised block experiments
• factorial experiments and interactions
• nested designs and repeated measures designs

10 credits

This module aims to provide you with a range of applied statistical techniques that can be used in professional life to analyse multivariate data. Both statistical and machine learning approaches are included.

You'll cover topics such as:

• graphical methods for investigating multivariate data
• logistic regression and discrimination
• linear and quadratic discriminant analysis
• non-parametric classification
• hierarchical and non-hierarchical clustering
• principal component analysis

20 credits

This module will cover the fundamental statistical methods necessary for the application of classical statistical methods to data collected for health care research. There will be an emphasis on the use of real data and the interpretation of statistical analyses in the context of the research hypothesis under investigation.

Topics covered will include:

• survival analysis
• analysing categorical data using hypothesis tests
• experimental Design and sampling
• clinical measurement

60 credits

The aim of this project is to develop research, teamwork, communication and time management skills, as well as collate evidence of your competency as a medical statistician. The class will also allow you to learn to critically appraise written work, provide constructive feedback, problem solve using a range of analytic tools, and deepen your knowledge and understanding of an area of research. 

10 credits

Most people have an intuitive understanding of what risk is. The aim of this module is to formalise this understanding and develop models to quantify risk. Quantification of risk relies on many statistical methods. The emphasis in this course is the practical use of such methods.

You'll develop skills in communicating risk to risk managers as well as formulating practical risk questions that can influence policy decisions.

You can expect to learn about:

• the difference between uncertainty and variability
• quantifying uncertainty using methods such as bootstrapping and Bayesian inference
• selecting appropriate probability distributions based on given scenarios
• fitting probability distributions to data
• building risk models
• communicating your results as written reports

All theory will be implemented practically via computing sessions using the statistical software R. You'll learn to create bespoke functions in R to implement your models and use summary statistics and plots to communicate your results. 

10 credits

This module will introduce you to Bayesian statistics and the modern Bayesian methods that are used in a variety of applications. Like with other modules, the focus is on real-life data and using statistical software packages for analysis.
You will gain experience in working with the following:

• visualising spatial data
• geospatial data, including methods for prediction
• bayesian modelling using software to implement Markov Chain Monte Carlo
• areal unit modelling

10 credits

Surveys are an important way of collecting data. This module will introduce you to the methods that are commonly used in health care to design questionnaires and analyse data resulting from these questionnaires.

You will consider:

• how to design appropriate survey questions
• a variety of sampling methods
• analysing data for different sampling methods

10 credits

This module will develop your skills in data presentation and statistical communication. You will learn to develop data dashboards, which are increasingly used to allow key stakeholders (and the public) to gain key insights into data via interactive visualisation.

Topics covered will include:

• Creating a data dashboard in RStudio
• User interface design with respect to accessibility
• Creating interactive data visualisations which reflect a specific aim
• Reactive programming  in RStudio
• Static programming in R

10 credits

This module will introduce the challenges of analysing big data with specific focus on the algorithms and techniques which are embodied in data analytics solutions.

At the end of the module, you'll understand:

• the fundamentals of Python for use in big data technologies
• how classical statistical techniques are applied in modern data analysis
• the limitations of various data analysis tools in a variety of contexts

10 credits

This module will enhance your understanding of the challenges posed by the advent of Big Data and will introduce you to scalable solutions for data storage and usage.

You can expect to learn about:

• the design and implementation of cloud NoSQL systems
• addressing design trade-offs and their impact
• the Map-Reduce programming paradigm

10 credits

This module provides you with the basic theories of machine learning and how to construct a machine model for a real dataset using R. You will also understand the ethical issues regarding data processing and management.

On completion of this module, you will be able to:

• clean data using RStudio and the tidyverse
• understand missing data and the role it plays
• understand ethical issues regarding data processing and management
• carry out single-value imputation
• carry out multiple imputed chained equations in R
• understand and implement artificial neural networks
• understand and implement support vector machines
• understand and implement tree-based classification and regression techniques
• understand and implement ensemble methods

10 credits

This module covers all aspects of statistical consultancy skills necessary for being a successful statistician working in any research or customer environment. You will work on real-life problems in small groups and have the opportunity to interact with stakeholders researchers to formulate hypotheses.

This module will cover how to:

• engage with professionals working in business, industry and the public sector
• apply their statistical knowledge in different situations
• effectively communicate statistical results to non-statisticians

10 credits

This module introduces deep Learning prediction algorithms covering background theory and practical application in Python.

Topics covered include:

• activation functions
• gradient descent, backpropagation and optimisation
• convolutional Neural Networks
• recurrent Neural Networks
• generative Adversarial Networks