Do you want to help your organisation improve productivity and enhance performance? A highly developed understanding of statistical models will do just that.
Our Advanced Statistical Analysis course has been designed to help those with a background in statistics to understand and use more advanced statistical models to make better sense of their data.
Through highly interactive workshops, gain a better understanding of the concepts behind advanced statistics; learn how to apply the Bayesian model and utilise more complex statistical methods, such as non-linear curves and hypothesis testing.
Practice using R programming on public sector data sets, which will help you utilise more advance statistical methods to improve your organisation’s performance.
Robert is a trainer, coach and writer on statistics and working in data science, especially data visualisation and Bayesian models.
He taught statistics and research methods to postgraduate clinical research students at St George’s Medical School and Kingston University (2010-2017), and contributed to many health services and biomedical research projects in this time. His freelance clients include Harvard Medical School, The Economist, and the Cabinet Office.
He is a fellow of the Royal Statistical Society and served on their statistical computing committee from 2012-16. He worked on clinical audits, analysing hospital quality and safety ...
Trainer’s Welcome and Clarification of Learning Objectives
Key Concepts behind Advanced Statistics
- The Data Generating Process
- Introduction to R programming
- Review of randomness and trends
- Establish what type of data you should be using
- Examine real-life public-sector statistics and data
- Gain an overview of machine learning: predictive analysis, AI and cloud computing
Applying the Bayesian Theory
- Missing data
- Known biases
- Intuitive probability-based outputs
- Incorporating prior evidence or opinion
- Establish areas in modelling and imperfect data where the Bayesian theory can help
Workshop: Utilising Advanced Statistical Methods Part 1
- Non-linear curves
- Hypothesis testing
- Confidence intervals
- Predictive distributions
- Multilevel models for clustered data
- Models for different outcomes (binary events or counts)
- Identify different methods in which statistical models can be used
- Trainer will review the day’s learning and the next stages of the course
- Delegates will have time to ask questions and share views with one another
Workshop: Utilising Advanced Statistical Methods Part 2
- Helpful assumptions
- The bias/variance trade-off
- Models with more than one predictor
- Interactions and non-linear functions of the predictors
- Constructing such a model for cause and effect or prediction