29 May Casual inference from observational data: instrumental variables (online)
1:00 pm - 2:30 pm
statistics course | level: intermediate |
affiliation: Ghent University
Are you concerned about how reproducible your data derived results will be? Have you seen statistics in a class but want to dig deeper? Are you using a statistical method but wonder if it is the best one to use?
The editors of Nature appreciate that it is tricky enough for a scientist to keep up-to-date in their own field, let alone in the ever expanding field of statistics. To help ease the burden on scientists, they have introduced a column on statistics to one of their publications, Nature Methods, called Points of Significance.
On regular occasions during the academic year, a statistician from FIRE or FLAMES will lead a discussion of a statistics topic from a Points of Significance article. We will start with the basics, that is, with the idea of sampling, work our way through a detailed discussion of ANOVA and end with the topic of Bayesian statistics.
This seminar is about 'Causal Inference from Observational Data: Instrumental Variables':
We have previously presented methods to mitigate the impact of measured confounding on the estimation of a causal treatment effects such as Matching and Weighting using propensity scores. A main critic of these methods is that they only control for measured confounders and do not control for unmeasured confounders. Instrumental variables analysis is a method for controlling for unmeasured confounding. This type of analysis requires the measurement of a valid instrumental variable, which is a variable that (i) is independent of the unmeasured confounding; (ii) affects the treatment; and (iii) affects the outcome only indirectly through its effect on the treatment. In summary, the IV method seeks to find a randomized experiment embedded in an observational study and uses this to estimate the causal treatment effect.
This tutorial discusses the types of causal effects that can be estimated by instrumental variables analysis; the assumptions; methods of estimation and sources of instrumental variables.
We will wrap up with an example that uses the ivpack package in the R software.
Baiocchi, Michael et al. “Instrumental variable methods for causal inference.” Statistics in medicine vol. 33,13 (2014): 2297-340. doi:10.1002/sim.6128
Dr Emmanuel Abatih