Introduction to interrupted time series analysis (ITSA) and segmented regression
ARC OxTV Training & Development Training Workshop
Wednesday, 09 March 2022 to Thursday, 10 March 2022
Maplethorpe Seminar Room, St High’s College, Oxford
Hosted by Nuffield Department of Primary Care Health Sciences
This course will serve as an introduction to interrupted time series analysis (ITSA) including selection and setup of data sources, statistical analysis, interpretation and presentation, and identification of potential challenges. ***Free to staff working within health, social care, public health and the ARC within the Oxford and Thames Valley region***
Organisers: Professor Richard Stevens, Dr Catia Nicodemo, Nicola Pidduck (NDPCHS University of Oxford)
***Free to staff working within health, social care, public health and the ARC within the Oxford and Thames Valley region***
This introductory course is available as a one-day or a two-day course. The one-day course is aimed at analysts and decision-makers wishing to understand and appraise evidence of this type. The two-day course is aimed at those wishing to understand and carry out simple interrupted time series, for example data analysts statisticians.
The course will cover: Introduction to linear regression models; use of segmented regression to evaluate interventions; issues arising when using regression in time series (autocorrelation, heteroskedasticity, seasonality). The one-day course combines theory with case studies and examples.
The two-day course extends this with an additional day of computer practical classes. Attendees of the two-day course should bring a laptop with the open-source software R (freely available from https://www.r-project.org/) for the second day.
Day 1 overview:
- How interventions are evaluated with segmented linear regression
- Special issues in time series data
- Introduction to statistical computing in R
- Interrupted time series analysis in R
To register, click here.
For further information please contact firstname.lastname@example.org