Royal Statistical Society
Highlands Local Group
Local organising Committee
We are pleased to announce that Brennan Kahan, Senior Research Fellow at the MRC Clinical Trials Unit at UCL, will deliver a talk on ‘Estimands in randomised trials: rethinking common wisdom’ on Thurs 30th September 2021.
The talk will be presented online over Microsoft Teams and details on registration will be advertised in due course. Further details, including the abstract, are below:
Date: Thurs 30th September 2021
Speaker: Brennan Kahan, Senior Research Fellow at the MRC Clinical Trials Unit at UCL
Title: Estimands in randomised trials: rethinking common wisdom
Summary: The interpretation of estimated treatment effects in randomised trials is often unclear or opaque. Estimands are a way to solve this issue, through a precise specification of what the treatment effect represents. In this talk, I will discuss how estimands can be used to ensure trials are addressing clinically relevant questions, and how they sometimes make us rethink conventional wisdom around which statistical methods are appropriate.
We are pleased to announce that Dr Tim Morris, statistician at the MRC Clinical Trials Unit at UCL, will deliver a talk on ‘Understanding performance measures in simulation studies’ on Wed 21st April 2021.
The talk will be presented online over Microsoft Teams and details on registration will be sent around in due course. Further details, including the abstract, are below:
Date: Wed 21st April 2021
Speaker: Tim Morris, MRC Clinical Trials Unit, UCL
Title: Understanding performance measures in simulation studies
Abstract: Statisticians use simulation studies to evaluate the performance of statistical methods. Performance measures are what we use to evaluate performance; examples include bias and coverage. This talk will describe some of the key performance measures that can be used. I will consider simulation studies of statistical methods that output a point estimate, standard error and (possibly) confidence interval. I will describe several performance measures by outlining what they aim to quantify, how they are calculated, and how Monte Carlo standard errors are estimated. Performance measures may use the (repeated) point estimates, the standard errors, the confidence intervals or more than one of these. Understanding performance measures and how they link together can help us to better critique simulation studies that we read and better plan our own. A key message is that it is rarely appropriate to evaluate only one performance measure.
Previous Seminars (from May
Related information of interest to fellows
Links to the Royal Statistical Society
Website maintained by