Welcome to the LSHTM Centre for Statistical Methodology
The LSHTM Centre for Statistical Methodology aims to gather statistical and methodological expertise across the School in order to strengthen our research capacity in epidemiology and public health, and exploit the richness and interdisciplinarity of the work carried out within the School. This is in line with the School’s strategic plans for sustaining areas of established expertise in which we have an international reputation.Read more
Next CSM Seminar
The Centre hosts regular seminars across the different themes. The next seminar is:Thurs 25 April 2019, 12:45-2:00pm LG8, Keppel Street Anders Huitfeldt (LSE) Title: "A new approach to generalizability of clinical trials" VanderWeele (Epidemiologic Methods, 2012) provided two separate definitions of effect heterogeneity, which he referred to as "effect modification in distribution" and "effect modification in measure". The standard epidemiological approach, which is based on effect modification in measure, is associated with a number of well-described shortcomings, and no consensus exists about the conditions under which investigators can assume effect homogeneity on either the additive or the multiplicative scale. More recently, Bareinboim and Pearl introduced a new graphical framework for transportability, based on effect heterogeneity in distribution. These graphs are an elegant solution to many of the problems associated with traditional approaches, but they require the investigator to make strong assumptions about the data generating mechanism: In particular, it is not sufficient to control for those variables that are associated with the effect of treatment; investigators using this approach are required to account for all causes of the outcome that differ between the populations. In light of these limitations, we propose a new definition of effect heterogeneity, based on “counterfactual outcome state transition parameters”, that is, the proportion of those individuals who would not have been a case by the end of follow-up if untreated, who would have responded to treatment by becoming a case; and the proportion of those individuals who would have become a case by the end of follow-up if untreated who would have responded to treatment by not becoming a case. Effects are said to be equal between populations if and only if these proportions are equal between the populations. Although counterfactual outcome state transition parameters are generally not identified from the data without strong monotonicity assumptions, we show that when they stay constant between populations, there are important implications for model specification, meta-analysis, and research generalization. Read more
Recent seminar slides & audio
Slides and audio recordings of many of our previous seminars are available. Our most recent seminars have been:Causal Inference Theme
Using Quantitative Bias Analysis to Deal with Misclassification in the Results Section, not the Discussion Section.
Matt Fox (Boston University)
Statistical Computing Theme
An extended mixed-effects model for meta-analysis: statistical framework and the R package mixmeta.
Antonio Gasparrini and Francesco Sera (LSHTM)
Big Data Theme
Large numbers of explanatory variables.
Heather Battey (Imperial College London)
Clinical Trials Theme
Design and analysis of trials where the outcome is a rate of change, with an introduction to a new Stata package for sample size calculation
Chris Frost and Amy Mullick (LSHTM)
Missing Data Theme
Uncertainty and missing data in dietary intake and activity data.
Graham Horgan (Rowett Institute, University of Aberdeen)
Centre for Statistical Methodology Mini-Symposium: Landmarking for survival analysis
Friday 18th January 2019
LSHTM, Keppel Street
Sequential Trials Symposium
The Centre for Statistical Methodology and the Tropical Epidemiology Group of the London School of Hygiene & Tropical Medicine held a half day symposium on sequential trials on the afternoon of 26 September 2018.Read more
"Statistical Methods for Big Data” Symposium
The Centre for Statistical Methodology held a half-day symposium on Statistical Methods for Big Data on 7 July 2017. The symposium aimed to discuss features of Big Data and the challenges that they pose for statistical methods and future directions of research. Speakers from different methodological perspectives presented examples across a wide spectrum of applications.Read more